Welcome to Backtesting tool comparison’s documentation!

Are you a newbie programmer or you are newbie trader ? This is your tutorial, where you can find some samples of code for downloading of historical data and for backtesting some samples of strategies.

And you may also find some interesting details in the Comparison and Conclusion sections.

Getting started

These contents can help you to start with your quantitative trading (quant trading or QT) system: you can find some samples for the main steps.

There are

  • many platforms can give you historical data

  • a lot of program languages that you can use for implementing your strategies

  • many trading systems that you can use for backtesting your system

Read the documentation on readthedocs for

  • a sample of the main backtesting steps for a few of the main languages are used for statistical analysis, graphics representation and reporting

  • comparison among languages and methods for backtesting your strategies

  • the best practices for your quant trading system

Goals

These contents can be useful at the beginning, when you are newbie programmer.

And it can help you to evaluate which data, language or trading system you need.

Disclaimer

The strategies contained in this tutorial, are some simple samples for having an idea how to use the libraries: those strategies are for the educational purpose only. All investments and trading in the stock market involve risk: any decisions related to buying/selling of stocks or other financial instruments should only be made after a thorough research, backtesting, running in demo and seeking a professional assistance if required.

Contribution

The documentation for R and Python languages, it has been powered by Jupyter:

$ git clone https://github.com/bilardi/backtesting-tool-comparison
$ cd backtesting-tool-comparison/
$ pip install --upgrade -r requirements.txt
$ docker run --rm -p 8888:8888 -e JUPYTER_ENABLE_LAB=yes -v "$PWD":/home/jovyan/ jupyter/datascience-notebook

The images are hosted on S3 and not in this repository:

  • use a PR for sharing a new version without images, only the new url

  • it will be our care to move them to S3 with all the others

For testing on your local client the documentation, see this README.md file.

License

These contents are released under the MIT license. See LICENSE for details.

R

Prerequisites

For using the program language R, you have to install r.

For mac,

$ brew install r

There are a lot of guide for getting started with R: knowing the basics will be taken for granted. For installing some packages, you can use RStudio or directly console

$ r
# general tools
install.packages("httr")
install.packages("jsonlite")
install.packages("plotly")
install.packages("devtools")
# for downloading data
install.packages("quantmod") # for Yahoo finance historical data of stock market
install.packages("Quandl")
devtools::install_github("amrrs/coinmarketcapr") # for Crypto currencies
devtools::install_github("amrrs/coindeskr") # for Crypto currencies
# for processing the data: in addition to Quantmod also QuantStrat
devtools::install_github("braverock/blotter")
devtools::install_github("braverock/quantstrat")

If you want to use Jupyter, you can use directly the commands below

$ git clone https://github.com/bilardi/backtesting-tool-comparison
$ cd backtesting-tool-comparison/
$ docker run --rm -p 8888:8888 -e JUPYTER_ENABLE_LAB=yes -v "$PWD":/home/jovyan/ jupyter/r-notebook

Jupyter is very easy for having a GUI virtual environment: knowing the basics will be taken for granted.

You can find all scripts descripted in this tutorial on GitHub:

  • the .r files in src/r/ folder, the script that you can use on RStudio

  • the .ipynb files in docs/sources/r/ folder, that you can use on Jupyter or browse on this tutorial

Getting started

R is a programming language and software environment for statistical analysis, graphics representation and reporting.

This is a summary of the r language syntax: you can find more details in cran documentation.

Basics

[1]:

# initialization of a variable
my_numeric_variable <- 1
my_string_variable <- "hello"
# print a variable
my_numeric_variable # and type enter
print(my_string_variable)
1
[1] "hello"

Vectors

[2]:

# initialization of a vector
my_numeric_vector <- c(1,2,3,4,5,6)
my_sequence <- 1:25
my_string_vector <- c("hello", "world", "!")
my_logic_vector <- c(TRUE, FALSE, T, F)
assign("my_vector", c(1:6))
my_random_vector <- sample(1:5000, 25) # Vector with 25 elements with random number from 1 to 5000

[3]:

# operations with vectors
sum(my_numeric_vector)
mean(my_numeric_vector)
median(my_numeric_vector)
21
3.5
3.5
[4]:

# multiplication 2 with each element of a vector
my_vector*2
# division 2 with each element of a vector
my_vector/2
  1. 2
  2. 4
  3. 6
  4. 8
  5. 10
  6. 12
  1. 0.5
  2. 1
  3. 1.5
  4. 2
  5. 2.5
  6. 3
[5]:

# sum each element of a vector with another vector by position
my_new_vector <- my_numeric_vector + my_vector # allow
my_new_vector <- my_numeric_vector + my_sequence # allow but with warning
my_new_vector <- my_numeric_vector + c(1:5) # allow but with warning

Warning message in my_numeric_vector + my_sequence:
“longer object length is not a multiple of shorter object length”
Warning message in my_numeric_vector + c(1:5):
“longer object length is not a multiple of shorter object length”

[6]:

# get element from a vector
my_string_vector[1] # print hello
my_string_vector[-2] # print hello!
my_string_vector[1:2] # print hello world
my_string_vector[c(1,3)] # print hello!
'hello'
  1. 'hello'
  2. '!'
  1. 'hello'
  2. 'world'
  1. 'hello'
  2. '!'
[7]:

# add labels to a vector
names(my_vector) <- c("one","two","three","four","five","six")
print(my_vector)

  one   two three  four  five   six
    1     2     3     4     5     6

[8]:

# get data type of a vector
class(my_vector) # print integer
typeof(my_string_vector) # print character
mode(my_vector) # print numeric
'integer'
'character'
'numeric'
[9]:

# convertion of each element of a vector
as.character(my_vector)
  1. '1'
  2. '2'
  3. '3'
  4. '4'
  5. '5'
  6. '6'

Matrixes

A matrix is a vector with more dimensions

[10]:

# matrix
my_matrix <- matrix(1:10, nrow = 2, ncol = 5) # creates a matrix bidimensional with 2 rows and 5 columns
my_matrix
my_matrix[2,2] # prints 4
my_matrix[1,] # prints row 1
my_matrix[,2] # prints column 2
A matrix: 2 × 5 of type int
1357 9
246810
4
  1. 1
  2. 3
  3. 5
  4. 7
  5. 9
  1. 3
  2. 4
[11]:

# merge of vectors
vector_one <- c("one", 0.1)
vector_two <- c("two", 1)
vector_three <- c("three", 10)
my_vectors <- matrix(c(vector_one, vector_two, vector_three), nrow = 3, ncol = 2, byrow = 1)
colnames(my_vectors) <- c("vector number", "quantity")
rownames(my_vectors) <- c("yesterday", "today", "tomorrow")

[12]:

my_vectors
A matrix: 3 × 2 of type chr
vector numberquantity
yesterdayone 0.1
todaytwo 1
tomorrowthree10

Weighted average

[13]:

# performance
apple_performance <- 3
netflix_performance <- 7
amazon_performance <- 11
# weight
apple_weight <- .3
netflix_weight <- .4
amazon_weight <- .3
# portfolio performance
weighted_average <- apple_performance * apple_weight + netflix_performance * netflix_weight + amazon_performance * amazon_weight
weighted_average
7
[14]:

# the same sample but with vectors
performance <- c(3,7,11)
weight <- c(.3,.4,.3)
company <- c('apple','netflix','amazon')
names(performance) <- company
names(weight) <- company
performance_weight <- performance * weight
performance_weight
weighted_average <- sum(performance_weight)
weighted_average
apple
0.9
netflix
2.8
amazon
3.3
7

Functions

[15]:

# the weighted average but with a function
weighted_average_function <- function(performance, weight) {
    return(sum(performance * weight))
}
performance <- c(3,7,11)
weight <- c(.3,.4,.3)
weighted_average <- weighted_average_function(performance, weight)
weighted_average
7

Historical data

This page contains some samples for downloading historical data of your quant trading system.

Quantmod library

Stock markets from Yahoo finance by Quantmod library (doc)

[1]:

library(quantmod)
symbol <- "AMZN"
getSymbols(symbol) # or getSymbols(symbol, src = "yahoo")

Loading required package: xts

Loading required package: zoo


Attaching package: ‘zoo’


The following objects are masked from ‘package:base’:

    as.Date, as.Date.numeric


Loading required package: TTR

Registered S3 method overwritten by 'quantmod':
  method            from
  as.zoo.data.frame zoo

‘getSymbols’ currently uses auto.assign=TRUE by default, but will
use auto.assign=FALSE in 0.5-0. You will still be able to use
‘loadSymbols’ to automatically load data. getOption("getSymbols.env")
and getOption("getSymbols.auto.assign") will still be checked for
alternate defaults.

This message is shown once per session and may be disabled by setting
options("getSymbols.warning4.0"=FALSE). See ?getSymbols for details.
'AMZN'
[2]:

head(AMZN) # returns some data of AMZN vector

           AMZN.Open AMZN.High AMZN.Low AMZN.Close AMZN.Volume AMZN.Adjusted
2007-01-03     38.68     39.06    38.05      38.70    12405100         38.70
2007-01-04     38.59     39.14    38.26      38.90     6318400         38.90
2007-01-05     38.72     38.79    37.60      38.37     6619700         38.37
2007-01-08     38.22     38.31    37.17      37.50     6783000         37.50
2007-01-09     37.60     38.06    37.34      37.78     5703000         37.78
2007-01-10     37.49     37.70    37.07      37.15     6527500         37.15

[3]:

colnames(AMZN) # returns column names of AMZN vector
  1. 'AMZN.Open'
  2. 'AMZN.High'
  3. 'AMZN.Low'
  4. 'AMZN.Close'
  5. 'AMZN.Volume'
  6. 'AMZN.Adjusted'
[4]:

#Op(AMZN) # only column of open of all data
#Hi(AMZN) # only column of high of all data
#Lo(AMZN) # only column of low of all data
#Cl(AMZN) # only column of close of all data
#Vo(AMZN) # only column of volume of all data
#Ad(AMZN) # value adjusted of the close (of all data)
#OHLC(AMZN) # columns open, high, low and close of all data
Exchange rates (forex) from Yahoo finance
[5]:

library(quantmod)
cross_from <- c("EUR", "USD", "CAD")
cross_to <- c("USD", "JPY", "USD")
getQuote(paste0(cross_from, cross_to, "=X"))
A data.frame: 3 × 8
Trade TimeLastChange% ChangeOpenHighLowVolume
<dttm><dbl><dbl><dbl><dbl><dbl><dbl><int>
EURUSD=X2021-01-11 12:21:43 1.2162491-0.006693244-0.5473087 1.2226434 1.2239902 1.21580550
USDJPY=X2021-01-11 12:22:10104.2400000 0.291000370 0.2799453103.9490000104.2490000103.68700000
CADUSD=X2021-01-11 12:22:10 0.7822278-0.006048977-0.7673632 0.7882768 0.7886435 0.78220950
Exchange rates (forex) from Oanda
[6]:

library(quantmod)
currency <- "EUR/GBP"
# the last 6 months
getSymbols(currency, src = "oanda")
head(EURGBP) # returns some data of EURGBP vector
'EUR/GBP'
            EUR.GBP
2020-07-16 0.907685
2020-07-17 0.909031
2020-07-18 0.909320
2020-07-19 0.909368
2020-07-20 0.907640
2020-07-21 0.903154

[7]:

# the last 60 days
getSymbols(currency, from = Sys.Date() - 60, to = Sys.Date(), src = "oanda")
head(EURGBP) # returns some data of EURGBP vector
'EUR/GBP'
            EUR.GBP
2020-11-12 0.895798
2020-11-13 0.898187
2020-11-14 0.896580
2020-11-15 0.896644
2020-11-16 0.897208
2020-11-17 0.896298

CoinMarketCap API

Crypto currencies by coinmarketcap (doc) and coindeskr (doc) Many libraries are not supported for a long period or it may no longer be supported because it has violated a policy:

  • the library coindeskr has violated the CRAN’s policy and it has archived on 2020-09-04

  • its GitHub repository is already available, but it has not updated for a change of CoinMarketCap API

[8]:

library(coinmarketcapr)
library(coindeskr)
last_marketcap <- get_global_marketcap("EUR")
today_marketcap <- get_marketcap_ticker_all()
head(today_marketcap[,1:8])
last_month <- get_last31days_price() # default it is Bitcoin
current_price <- get_current_price() # default it is Bitcoin

⚠ The old API is used when no 'apikey' is given.

Error: lexical error: invalid char in json text.
                                       <!DOCTYPE HTML> <html lang="en-
                     (right here) ------^

Traceback:

1. get_global_marketcap("EUR")
2. check_response(d)
3. stop(cat(crayon::red(cli::symbol$cross, "The request was not succesfull! \n",
 .     "Request URL:\n", req$url, "\n", "Response Content:\n", jsonlite::prettify(rawToChar(req$content)),
 .     "\n")))
4. cat(crayon::red(cli::symbol$cross, "The request was not succesfull! \n",
 .     "Request URL:\n", req$url, "\n", "Response Content:\n", jsonlite::prettify(rawToChar(req$content)),
 .     "\n"))
5. crayon::red(cli::symbol$cross, "The request was not succesfull! \n",
 .     "Request URL:\n", req$url, "\n", "Response Content:\n", jsonlite::prettify(rawToChar(req$content)),
 .     "\n")
6. mypaste(...)
7. lapply(list(...), as.character)
8. jsonlite::prettify(rawToChar(req$content))
9. reformat(txt, TRUE, indent_string = paste(rep(" ", as.integer(indent)),
 .     collapse = ""))
10. stop(out[[2]], call. = FALSE)
11. base::stop(..., call. = FALSE)

Below you can find a sample with sandbox API of CoinMarketCap.

[9]:

library(httr)
library(jsonlite)
base_url <- 'https://sandbox-api.coinmarketcap.com/'

get_coinmarketcap_data <- function(base_url, path, query) {
    response <- GET(url = base_url, path = paste('v1/', path, sep = ''), query = query)
    content <- content(response, 'text', encoding = 'utf-8')
    return(fromJSON(content, flatten = TRUE))
}

# returns all active cryptocurrencies
path <- 'cryptocurrency/map'
marketcap_map <- get_coinmarketcap_data(base_url, path, list())
head(marketcap_map$data)
A data.frame: 6 × 11
idnamesymbolslugis_activerankplatform.idplatform.nameplatform.symbolplatform.slugplatform.token_address
<int><chr><chr><chr><int><int><int><chr><chr><chr><chr>
11Bitcoin BTCbitcoin 1 1NANANANANA
22Litecoin LTClitecoin 1 5NANANANANA
33Namecoin NMCnamecoin 1310NANANANANA
44TerracoinTRCterracoin1926NANANANANA
55Peercoin PPCpeercoin 1346NANANANANA
66Novacoin NVCnovacoin 1842NANANANANA
 
[10]:

# returns last market data of all active cryptocurrencies
path <- 'cryptocurrency/listings/latest'
query <- list(convert = 'EUR')
last_marketcap <- get_coinmarketcap_data(base_url, path, query)
head(last_marketcap$data)
A data.frame: 6 × 24
idnamesymbolslugnum_market_pairsdate_addedtagsmax_supplycirculating_supplytotal_supply⋯platform.symbolplatform.slugplatform.token_addressquote.EUR.pricequote.EUR.volume_24hquote.EUR.percent_change_1hquote.EUR.percent_change_24hquote.EUR.percent_change_7dquote.EUR.market_capquote.EUR.last_updated
<int><chr><chr><chr><int><chr><list><dbl><dbl><dbl>⋯<chr><chr><chr><dbl><dbl><lgl><lgl><lgl><dbl><chr>
1 1Bitcoin BTC bitcoin 79192013-04-28T00:00:00.000Zmineable2.1e+07 17906012 17906012⋯NA NA NA8708.442730312507935648NANANA1559334800302019-08-30T18:51:01.000Z
21027Ethereum ETH ethereum 56292015-08-07T00:00:00.000Zmineable NA 107537936 107537936⋯NA NA NA 153.6859725 5260772807NANANA 165270723362019-08-30T18:51:01.000Z
3 52XRP XRP ripple 4492013-08-04T00:00:00.000Z1.0e+114293286696799991366793⋯NA NA NA 0.2318186 844359719NANANA 99526384542019-08-30T18:51:01.000Z
41831Bitcoin CashBCH bitcoin-cash 3782017-07-23T00:00:00.000Zmineable2.1e+07 17975975 17975975⋯NA NA NA 256.0351264 1281819319NANANA 46024810312019-08-30T18:51:01.000Z
5 2Litecoin LTC litecoin 5382013-04-28T00:00:00.000Zmineable8.4e+07 63147124 63147124⋯NA NA NA 58.6260748 2207389317NANANA 37020680162019-08-30T18:51:01.000Z
6 825Tether USDTtether 30162015-02-25T00:00:00.000Z NA 4008269411 4095057493⋯OMNIomni31 0.913871414536357084NANANA 36630427622019-08-30T18:51:01.000Z
 
[11]:

# returns last market data of Bitcoin
path <- 'cryptocurrency/quotes/latest'
query <- list(id = 1, convert = 'EUR')
last_bitcoin <- get_coinmarketcap_data(base_url, path, query)
last_bitcoin$data$`1`
last_bitcoin$data$`1`$`quote`$EUR
$id
1
$name
'Bitcoin'
$symbol
'BTC'
$slug
'bitcoin'
$num_market_pairs
7919
$date_added
'2013-04-28T00:00:00.000Z'
$tags
'mineable'
$max_supply
21000000
$circulating_supply
17906012
$total_supply
17906012
$is_active
1
$is_market_cap_included_in_calc
1
$platform
NULL
$cmc_rank
1
$is_fiat
0
$last_updated
'2019-08-30T18:51:28.000Z'
$quote
$EUR =
$price
8708.44273026964
$volume_24h
12507935648.1974
$percent_change_1h
NULL
$percent_change_24h
NULL
$percent_change_7d
NULL
$market_cap
155933480029.521
$last_updated
'2019-08-30T18:51:01.000Z'
$price
8708.44273026964
$volume_24h
12507935648.1974
$percent_change_1h
NULL
$percent_change_24h
NULL
$percent_change_7d
NULL
$market_cap
155933480029.521
$last_updated
'2019-08-30T18:51:01.000Z'
[12]:

# returns historical data from a time_start to a time_end
path <- 'cryptocurrency/quotes/historical'
query <- list(id = 1, convert = 'EUR', time_start = '2020-05-21T12:14', time_end = '2020-05-21T12:44')
data <- get_coinmarketcap_data(base_url, path, query)
head(data$data)
$id
1
$name
'Bitcoin'
$symbol
'BTC'
$is_fiat
0
$quotes
A data.frame: 6 × 2
timestampquote.EUR.timestamp
<chr><chr>
12020-05-21T12:19:03.000Z2020-05-21T12:19:03.000Z
22020-05-21T12:24:02.000Z2020-05-21T12:24:02.000Z
32020-05-21T12:29:03.000Z2020-05-21T12:29:03.000Z
42020-05-21T12:34:03.000Z2020-05-21T12:34:03.000Z
52020-05-21T12:39:01.000Z2020-05-21T12:39:01.000Z
62020-05-21T12:44:02.000Z2020-05-21T12:44:02.000Z

Quandl library

OHLC (Open High Low Close) data by Quandl library (doc), for example crude oil

[13]:

library(Quandl)
oil_data <- Quandl(code = "CHRIS/CME_QM1")
head(oil_data) # returns all free columns names
A data.frame: 6 × 9
DateOpenHighLowLastChangeSettleVolumePrevious Day Open Interest
<date><dbl><dbl><dbl><dbl><dbl><dbl><dbl><dbl>
12021-01-0850.92552.75050.82552.750 1.4152.24107701833
22021-01-0750.52551.27550.40050.950 0.2050.83108141800
32021-01-0649.82550.92549.47550.525 0.7050.63166751908
42021-01-0547.40050.20047.27549.850 2.3149.93234731799
52021-01-0448.40049.85047.20047.350-0.9047.62218451350
62020-12-3148.32548.57547.77548.450 0.1248.52 66981446
 
[14]:

#getPrice(oil_data) # returns all free columns values
#getPrice(oil_data, symbol = "CHRIS/CME_QM1", prefer = "Open$") # returns open values of a specific symbol
getPrice(oil_data, prefer = "Open$") # returns open values
  1. 50.925
  2. 50.525
  3. 49.825
  4. 47.4
  5. 48.4
  6. 48.325
  7. 48.125
  8. 47.7
  9. 48.2
  10. 48.1
  11. 46.775
  12. 47.85
  13. 49.3
  14. 48.35
  15. 47.825
  16. 47.55
  17. 47.05
  18. 46.65
  19. 47
  20. 45.725
  21. 45.55
  22. 45.6
  23. 46.125
  24. 45.625
  25. 45
  26. 44.375
  27. 45.05
  28. 45.45
  29. 45.9
  30. 44.8
  31. 42.825
  32. 42.55
  33. 41.85
  34. 41.6
  35. 41.325
  36. 41.525
  37. 40.175
  38. 40.9
  39. 41.425
  40. 41.8
  41. 39.825
  42. 37.325
  43. 38.5
  44. 39.1
  45. 38.15
  46. 37.025
  47. 35.325
  48. 36.1
  49. 37.475
  50. 38.95
  51. 38.625
  52. 39.75
  53. 40.6
  54. 40.025
  55. 41.25
  56. 40.95
  57. 40.9
  58. 40.875
  59. 41.15
  60. 40.225
  61. 39.525
  62. 40.45
  63. 41.325
  64. 39.975
  65. 39.875
  66. 39.375
  67. 37
  68. 38.525
  69. 39.85
  70. 39.225
  71. 40.575
  72. 40.125
  73. 40.175
  74. 39.6
  75. 39.725
  76. 39.875
  77. 40.85
  78. 40.95
  79. 40.175
  80. 38.325
  81. 37.3
  82. 37.325
  83. 37.1
  84. 37.8
  85. 36.85
  86. 39.45
  87. 41.25
  88. 41.575
  89. 43.05
  90. 42.825
  91. 42.925
  92. 43
  93. 43.475
  94. 43.375
  95. 42.375
  96. 42.425
  97. 42.75
  98. 42.95
  99. 42.65
  100. 42.65
  101. 42.15
  102. 42.325
  103. 42.55
  104. 41.575
  105. 42
  106. 41.55
  107. 41.925
  108. 42.225
  109. 41.525
  110. 40.775
  111. 40.375
  112. 40.4
  113. 41.325
  114. 41.1
  115. 41.675
  116. 41.325
  117. 41.1
  118. 41.925
  119. 41.55
  120. 40.8
  121. 40.45
  122. 40.75
  123. 40.95
  124. 40.55
  125. 39.625
  126. 40.375
  127. 39.6
  128. 40.825
  129. 40.45
  130. 40.6
  131. 40.375
  132. 39.775
  133. 39.825
  134. 39.625
  135. 37.95
  136. 39.075
  137. 38.05
  138. 39.975
  139. 40.65
  140. 39.05
  141. 38.875
  142. 37.625
  143. 37.95
  144. 37.1
  145. 35.975
  146. 36.25
  147. 39.05
  148. 38.425
  149. 38.2
  150. 39.4
  151. 37.325
  152. 36.725
  153. 36.825
  154. 35.525
  155. 35.075
  156. 33.675
  157. 32.025
  158. 34.1
  159. 33.3
  160. 33.95
  161. 33.5
  162. 31.9
  163. 32.35
  164. 30
  165. 27.575
  166. 25.7
  167. 25.35
  168. 24.575
  169. 24.425
  170. 23.5
  171. 24
  172. 25.5
  173. 21.15
  174. 19.1
  175. 19.075
  176. 15.6
  177. 13.275
  178. 13
  179. 16.875
  180. 16.825
  181. 14.3
  182. 13
  183. 21.3
  184. 17.75
  185. 20.025
  186. 20.125
  187. 20.7
  188. 22.4
  189. 24.55
  190. 26.25
  191. 24.25
  192. 26.35
  193. 26.5
  194. 24.8
  195. 21.225
  196. 20.1
  197. 20.25
  198. 21
  199. 23.3
  200. 24.15
  201. ⋯
  202. 56.1
  203. 55.375
  204. 55.325
  205. 57.25
  206. 59.225
  207. 63.35
  208. 63
  209. 65.525
  210. 66.725
  211. 67.4
  212. 67.6
  213. 69.25
  214. 66.125
  215. 73.5
  216. 73.85
  217. 75.5
  218. 76.575
  219. 76.325
  220. 74.325
  221. 74.225
  222. 75.525
  223. 75.95
  224. 74.3
  225. 76.9
  226. 77.5
  227. 77.275
  228. 78.475
  229. 77.9
  230. 78.9
  231. 77.4
  232. 78.175
  233. 80.6
  234. 81
  235. 81.925
  236. 81.575
  237. 80.65
  238. 81.225
  239. 81.925
  240. 80.4
  241. 82.525
  242. 81.85
  243. 82.4
  244. 83.175
  245. 80.95
  246. 82.3
  247. 85.025
  248. 85.3
  249. 84.4
  250. 87.7
  251. 88.45
  252. 90.5
  253. 89.75
  254. 91.35
  255. 90.725
  256. 91.375
  257. 94.3
  258. 93.3
  259. 92.525
  260. 92.9
  261. 91.65
  262. 90.7
  263. 91.625
  264. 91.95
  265. 94
  266. 94.725
  267. 92.75
  268. 92.125
  269. 93.05
  270. 91.7
  271. 92.775
  272. 93.15
  273. 93.45
  274. 94.525
  275. 95.1
  276. 93.225
  277. 95.875
  278. 94.575
  279. 93.725
  280. 93.875
  281. 93.35
  282. 93.375
  283. 93.9
  284. 93.475
  285. 92.85
  286. 93.925
  287. 97.2
  288. 95.525
  289. 97.35
  290. 97.175
  291. 97.85
  292. 97.45
  293. 97.6
  294. 96.9
  295. 97.6
  296. 98.4
  297. 97.675
  298. 97.65
  299. 99.525
  300. 101
  301. 101.575
  302. 101.875
  303. 102.05
  304. 103.2
  305. 102
  306. 102.75
  307. 101.725
  308. 103.75
  309. 101.475
  310. 100.2
  311. 100.95
  312. 100.525
  313. 102.85
  314. 101.95
  315. 103.5
  316. 103.4
  317. 103.8
  318. 104.275
  319. 105.225
  320. 105.55
  321. 105.775
  322. 105.65
  323. 106.7
  324. 106.025
  325. 106
  326. 106.9
  327. 106.075
  328. 105.7
  329. 106.625
  330. 106.575
  331. 106.9
  332. 106.825
  333. 104.5
  334. 104.325
  335. 104.5
  336. 102.7
  337. 102.425
  338. 102.375
  339. 102.775
  340. 102.45
  341. 102.875
  342. 103.45
  343. 103.075
  344. 104.15
  345. 104.375
  346. 103.75
  347. 103.825
  348. 102.925
  349. 102.1
  350. 101.6
  351. 101.575
  352. 102
  353. 101.925
  354. 100.6
  355. 100.025
  356. 100.25
  357. 100.775
  358. 99.825
  359. 99.425
  360. 99.95
  361. 99.2
  362. 99.7
  363. 100.775
  364. 100.85
  365. 100.25
  366. 101.925
  367. 101.5
  368. 101.9
  369. 103.625
  370. 103.65
  371. 103.825
  372. 103.825
  373. 103.6
  374. 103.65
  375. 103.35
  376. 103.4
  377. 102.325
  378. 100.7
  379. 101
  380. 100.4
  381. 99.3
  382. 99.675
  383. 101.5
  384. 101.575
  385. 101.325
  386. 100.275
  387. 99.2
  388. 99.425
  389. 99.45
  390. 98.675
  391. 99.15
  392. 98.575
  393. 98.025
  394. 99.35
  395. 98.2
  396. 98.1
  397. 99.5
  398. 100.925
  399. 101.9
  400. 101
  401. 103.325

Graphics

This is a sample for plotting with R.

[1]:

library(quantmod)
# initialization
symbols = c("AMZN", "FB", "NFLX", "MSFT")
start <- as.Date("2017-01-01")
end <- as.Date("2020-01-01")
getSymbols(Symbols = symbols, src = "yahoo", from = start, to = end)
# get only Close values of AMZN and MSFT symbols
stocks <- as.xts(data.frame(AMZN = AMZN[,"AMZN.Close"], MSFT = MSFT[,"MSFT.Close"]))

Loading required package: xts

Loading required package: zoo


Attaching package: ‘zoo’


The following objects are masked from ‘package:base’:

    as.Date, as.Date.numeric


Loading required package: TTR

Registered S3 method overwritten by 'quantmod':
  method            from
  as.zoo.data.frame zoo

‘getSymbols’ currently uses auto.assign=TRUE by default, but will
use auto.assign=FALSE in 0.5-0. You will still be able to use
‘loadSymbols’ to automatically load data. getOption("getSymbols.env")
and getOption("getSymbols.auto.assign") will still be checked for
alternate defaults.

This message is shown once per session and may be disabled by setting
options("getSymbols.warning4.0"=FALSE). See ?getSymbols for details.
  1. 'AMZN'
  2. 'FB'
  3. 'NFLX'
  4. 'MSFT'
[2]:

# plotting
library(plotly)
plot(AMZN[,"AMZN.Close"], main = "AMZN") # prints linear graph

Loading required package: ggplot2


Attaching package: ‘plotly’


The following object is masked from ‘package:ggplot2’:

    last_plot


The following object is masked from ‘package:stats’:

    filter


The following object is masked from ‘package:graphics’:

    layout
_images/r_graphics_2_1.png 
[3]:

# create a custom theme
my_theme <- chart_theme()
my_theme$col$up.col <- "darkgreen"
my_theme$col$up.border <- "black"
my_theme$col$dn.col <- "darkred"
my_theme$col$dn.border <- "black"
my_theme$rylab <- FALSE
my_theme$col$grid <- "lightgrey"

[4]:

# using the custom theme with a range
chart_Series(AMZN, subset = "2017-10::2018-09", theme = my_theme)
_images/r_graphics_4_0.png 
[5]:

# using the custom theme with a range of one month
chart_Series(AMZN, subset = "2018-09", theme = my_theme)
_images/r_graphics_5_0.png 
[6]:

# using the black theme of quantmod
barChart(AMZN, theme = chartTheme('black'))
_images/r_graphics_6_0.png 
[7]:

# using the candle theme of quantmod
candleChart(AMZN, up.col = "green", dn.col = "red", theme = "white")
_images/r_graphics_7_0.png 
[8]:

# zoom of graph on one month
zoomChart("2018-09")
_images/r_graphics_8_0.png 
[9]:

# zoom of graph on one year
zoomChart("2018")
_images/r_graphics_9_0.png 
[10]:

# add indicators
addSMA(n = c(20, 50, 200)) # adds simple moving averages
_images/r_graphics_10_0.png 
[11]:

addBBands(n = 20, sd = 2, ma = "SMA", draw = "bands", on = -1) # sd = standard deviation, ma = average
_images/r_graphics_11_0.png 
[12]:

AMZN_ema_21 <- EMA(Cl(AMZN), n=21) # exponencial moving average
addTA(AMZN_ema_21, on = 1, col = "red")
_images/r_graphics_12_0.png 
[13]:

AMZN_ema_50 <- EMA(Cl(AMZN), n=50) # exponencial moving average
addTA(AMZN_ema_50, on = 1, col = "blue")
_images/r_graphics_13_0.png 
[14]:

AMZN_ema_200 <- EMA(Cl(AMZN), n=200) # exponencial moving average
addTA(AMZN_ema_200, on = 1, col = "orange")
_images/r_graphics_14_0.png 
[15]:

addRSI(n = 14, maType = "EMA", wilder = TRUE) # relative strength index
_images/r_graphics_15_0.png

Strategies

You can initilize a strategy by R language.

The main libraries are

How to load indicators on your strategy

[1]:

#install.packages("quantmod")
#require(devtools)
#devtools::install_github("braverock/blotter")
#devtools::install_github("braverock/quantstrat")

[2]:

library(quantmod)
# initialization
Sys.setenv(TZ="UTC")
symbol = as.character("AMZN")
start <- as.Date("2017-01-01")
end <- as.Date("2020-01-01")
getSymbols(Symbols = symbol, src = "yahoo", from = start, to = end)

Loading required package: xts

Loading required package: zoo


Attaching package: ‘zoo’


The following objects are masked from ‘package:base’:

    as.Date, as.Date.numeric


Loading required package: TTR

Registered S3 method overwritten by 'quantmod':
  method            from
  as.zoo.data.frame zoo

‘getSymbols’ currently uses auto.assign=TRUE by default, but will
use auto.assign=FALSE in 0.5-0. You will still be able to use
‘loadSymbols’ to automatically load data. getOption("getSymbols.env")
and getOption("getSymbols.auto.assign") will still be checked for
alternate defaults.

This message is shown once per session and may be disabled by setting
options("getSymbols.warning4.0"=FALSE). See ?getSymbols for details.
'AMZN'
[3]:

AMZN <- na.omit(lag(AMZN))
# calculate indicators
sma20 <- SMA(Cl(AMZN), n = 20)
ema20 <- EMA(Cl(AMZN), n = 20)

[4]:

library(quantstrat)
# strategy initialization
currency("USD")
stock(symbol, currency = 'USD', multiplier = 1)
tradeSize <- 100000
initEquity <- 100000
account.st <- portfolio.st <- strategy.st <- "my strategy"
rm.strat("my strategy")
out <- initPortf(portfolio.st, symbols = symbol, initDate = start, currency = "USD")
out <- initAcct(account.st, portfolios = portfolio.st, initDate = start, currency = "USD", initEq = initEquity)
initOrders(portfolio.st, initDate = start)
strategy(strategy.st, store = TRUE)

Loading required package: blotter

Loading required package: FinancialInstrument

Loading required package: PerformanceAnalytics


Attaching package: ‘PerformanceAnalytics’


The following object is masked from ‘package:graphics’:

    legend


Loading required package: foreach
'USD'
'AMZN'
[5]:

# indicators with QuantStrat
out <- add.indicator(strategy.st, name = "SMA", arguments = list(x = quote(Cl(AMZN)), n = 20, maType = "SMA"), label = "SMA20periods")
out <- add.indicator(strategy.st, name = "SMA", arguments = list(x = quote(Cl(AMZN)), n = 50, maType = "SMA"), label = "SMA50periods")
out <- add.indicator(strategy.st, name = "EMA", arguments = list(x = quote(Cl(AMZN)), n = 20, maType = "EMA"), label = "EMA20periods")
out <- add.indicator(strategy.st, name = "EMA", arguments = list(x = quote(Cl(AMZN)), n = 50, maType = "EMA"), label = "EMA50periods")
out <- add.indicator(strategy.st, name = "RSI", arguments = list(price = quote(Cl(AMZN)), n = 7), label = "RSI720periods")
out <- add.indicator(strategy.st, name = "BBands", arguments = list(HLC = quote(Cl(AMZN)), n = 20, maType = "SMA", sd = 2), label = "Bollinger Bands")

[6]:

# custom indicator
RSIaverage <- function(price, n1, n2) {
    RSI_1 <- RSI(price = price, n = n1)
    RSI_2 <- RSI(price = price, n = n2)
    calculatedAverage <- (RSI_1 + RSI_2) / 2
    colnames(calculatedAverage) <- "RSI_average"
    return(calculatedAverage)
}
out <- add.indicator(strategy.st, name = "RSIaverage", arguments = list(price = quote(Cl(AMZN)), n1 = 7, n2 = 14), label = "RSIaverage")

[7]:

# first tests
plot(RSIaverage(Cl(AMZN), n1=7, n2=14)) # for graphing your function results
_images/r_strategies_7_0.png 
[8]:

# load all indicators on my data
test <- applyIndicators(strategy = strategy.st, mktdata = OHLC(AMZN))
subsetTest <- test["2018-01-01/2019-01-01"]
head(subsetTest)

           AMZN.Open AMZN.High AMZN.Low AMZN.Close SMA.SMA20periods
2018-01-02   1182.35   1184.00  1167.50    1169.47         1168.841
2018-01-03   1172.00   1190.00  1170.51    1189.01         1170.174
2018-01-04   1188.30   1205.49  1188.30    1204.20         1173.687
2018-01-05   1205.00   1215.87  1204.66    1209.59         1177.088
2018-01-08   1217.51   1229.14  1210.00    1229.14         1180.927
2018-01-09   1236.00   1253.08  1232.03    1246.87         1185.281
           SMA.SMA50periods EMA.EMA20periods EMA.EMA50periods rsi.RSI720periods
2018-01-02         1129.739         1168.779         1130.025          44.92805
2018-01-03         1133.787         1170.706         1132.338          60.65982
2018-01-04         1138.213         1173.896         1135.156          68.75469
2018-01-05         1143.078         1177.295         1138.075          71.20734
2018-01-08         1148.143         1182.233         1141.646          78.38659
2018-01-09         1153.622         1188.389         1145.773          82.89832
           dn.Bollinger.Bands mavg.Bollinger.Bands up.Bollinger.Bands
2018-01-02           1140.397             1168.841           1197.286
2018-01-03           1140.596             1170.174           1199.753
2018-01-04           1145.498             1173.687           1201.876
2018-01-05           1148.806             1177.088           1205.370
2018-01-08           1146.863             1180.927           1214.992
2018-01-09           1142.097             1185.281           1228.466
           pctB.Bollinger.Bands RSI_average.RSIaverage
2018-01-02            0.5110476               49.34830
2018-01-03            0.8183989               60.54486
2018-01-04            1.0412143               66.72199
2018-01-05            1.0746036               68.64585
2018-01-08            1.2076629               74.50261
2018-01-09            1.2130873               78.45513

How to load signals on your strategy

The best practice is to prepare one column for each signal that you want to use on your strategy. The strategy below use the moving averages that they are SMA and EMA. You can use one or the other.

Disclaimer

The strategies below are some simple samples for having an idea how to use the libraries: those strategies are for the educational purpose only. All investments and trading in the stock market involve risk: any decisions related to buying/selling of stocks or other financial instruments should only be made after a thorough research, backtesting, running in demo and seeking a professional assistance if required.

Moving Average Crossover Strategy - Sample 1

  • when the price value crosses the MA value from below, it will close any existing short position and go long (buy) one unit of the asset

  • when the price value crosses the MA value from above, it will close any existing long position and go short (sell) one unit of the asset

Reference: https://www.learndatasci.com/tutorials/python-finance-part-3-moving-average-trading-strategy/

[9]:

# Moving Average Crossover Strategy - Sample 1
signalPosition <- function(price, ma) {
    # Taking the difference between the prices and the MA timeseries
    price_ma_diff <- price - ma
    price_ma_diff[is.na(price_ma_diff)] <- 0 # replace missing signals with no position
    # Taking the sign of the difference to determine whether the price or the MA is greater
    position <- diff(price_ma_diff)
    position[is.na(position)] <- 0 # replace missing signals with no position
    colnames(position) <- "Signal_position"
    return(position)
}

[10]:

#out <- add.indicator(strategy.st, name = "signalPosition", arguments = list(price = Cl(AMZN), ma = sma20), label = "signalPosition")
out <- add.indicator(strategy.st, name = "signalPosition", arguments = list(price = Cl(AMZN), ma = ema20), label = "signalPosition1")
out <- add.signal(strategy.st, name = "sigThreshold", arguments = list(column = "Signal_position", threshold = 2, relationship = "gte", cross = TRUE), label = "buy1")
out <- add.signal(strategy.st, name = "sigThreshold", arguments = list(column = "Signal_position", threshold = -2, relationship = "lte", cross = TRUE), label = "sell1")
#add.signal(strategy.st, name = "sigThreshold", arguments = list(data = position, column = "AMZN.Close", threshold = 2, relationship = "gte", cross = TRUE), label = "buy1")
#add.signal(strategy.st, name = "sigThreshold", arguments = list(data = position, column = "AMZN.Close", threshold = -2, relationship = "lte", cross = TRUE), label = "sell1")
# Creating rules
out <- add.rule(strategy.st, name = 'ruleSignal', arguments = list(sigcol = "buy1", sigval = TRUE, orderqty = 100, ordertype = 'market', orderside = 'long'), type = 'enter')
out <- add.rule(strategy.st, name = 'ruleSignal', arguments = list(sigcol = "sell1", sigval = TRUE, orderqty = 'all', ordertype = 'market', orderside = 'long'), type = 'exit')
#out <- add.rule(strategy.st, name = 'ruleSignal', arguments = list(sigcol = "sell1", sigval = TRUE, orderqty = -100, ordertype = 'market', orderside = 'short'), type = 'enter')
#out <- add.rule(strategy.st, name = 'ruleSignal', arguments = list(sigcol = "buy1", sigval = TRUE, orderqty = 'all', ordertype = 'market', orderside = 'short'), type = 'exit')

[11]:

# backtesting
applyStrategy(strategy.st, portfolios=portfolio.st)
out <- updatePortf(portfolio.st)
dateRange <- time(getPortfolio(portfolio.st)$summary)[-1]
out <- updateAcct(portfolio.st, dateRange)
out <- updateEndEq(account.st)
chart.Posn(portfolio.st, Symbol=symbol, TA=c("add_SMA(n=20,col='red')"))

[1] "2017-02-02 00:00:00 AMZN 100 @ 832.349976"
[1] "2017-02-07 00:00:00 AMZN -100 @ 807.640015"
[1] "2017-02-09 00:00:00 AMZN 100 @ 819.710022"
[1] "2017-02-14 00:00:00 AMZN 100 @ 836.530029"
[1] "2017-02-17 00:00:00 AMZN 100 @ 844.140015"
[1] "2017-02-23 00:00:00 AMZN 100 @ 855.609985"
[1] "2017-02-24 00:00:00 AMZN -400 @ 852.190002"
[1] "2017-03-01 00:00:00 AMZN 100 @ 845.039978"
[1] "2017-03-02 00:00:00 AMZN -100 @ 853.080017"
[1] "2017-03-03 00:00:00 AMZN 100 @ 848.909973"
[1] "2017-03-06 00:00:00 AMZN -100 @ 849.880005"
[1] "2017-03-10 00:00:00 AMZN 100 @ 853"
[1] "2017-03-16 00:00:00 AMZN -100 @ 852.969971"
[1] "2017-03-22 00:00:00 AMZN 100 @ 843.200012"
[1] "2017-03-23 00:00:00 AMZN -100 @ 848.059998"
[1] "2017-03-24 00:00:00 AMZN 100 @ 847.380005"
[1] "2017-03-30 00:00:00 AMZN 100 @ 874.320007"
[1] "2017-04-04 00:00:00 AMZN 100 @ 891.51001"
[1] "2017-04-06 00:00:00 AMZN 100 @ 909.280029"
[1] "2017-04-10 00:00:00 AMZN -400 @ 894.880005"
[1] "2017-04-12 00:00:00 AMZN 100 @ 902.359985"
[1] "2017-04-13 00:00:00 AMZN -100 @ 896.22998"
[1] "2017-04-19 00:00:00 AMZN 100 @ 903.780029"
[1] "2017-04-21 00:00:00 AMZN -100 @ 902.059998"
[1] "2017-04-26 00:00:00 AMZN 100 @ 907.619995"
[1] "2017-05-01 00:00:00 AMZN 100 @ 924.98999"
[1] "2017-05-04 00:00:00 AMZN -200 @ 941.030029"
[1] "2017-05-10 00:00:00 AMZN 100 @ 952.820007"
[1] "2017-05-12 00:00:00 AMZN -100 @ 947.619995"
[1] "2017-05-16 00:00:00 AMZN 100 @ 957.969971"
[1] "2017-05-17 00:00:00 AMZN -100 @ 966.070007"
[1] "2017-05-18 00:00:00 AMZN 100 @ 944.76001"
[1] "2017-05-19 00:00:00 AMZN -100 @ 958.48999"
[1] "2017-05-22 00:00:00 AMZN 100 @ 959.840027"
[1] "2017-05-24 00:00:00 AMZN 100 @ 971.539978"
[1] "2017-05-26 00:00:00 AMZN 100 @ 993.380005"
[1] "2017-06-01 00:00:00 AMZN -300 @ 994.619995"
[1] "2017-06-06 00:00:00 AMZN 100 @ 1011.340027"
[1] "2017-06-08 00:00:00 AMZN -100 @ 1010.070007"
[1] "2017-06-09 00:00:00 AMZN 100 @ 1010.27002"
[1] "2017-06-12 00:00:00 AMZN -100 @ 978.309998"
[1] "2017-06-15 00:00:00 AMZN 100 @ 976.469971"
[1] "2017-06-16 00:00:00 AMZN -100 @ 964.169983"
[1] "2017-06-20 00:00:00 AMZN 100 @ 995.169983"
[1] "2017-06-22 00:00:00 AMZN -100 @ 1002.22998"
[1] "2017-06-23 00:00:00 AMZN 100 @ 1001.299988"
[1] "2017-06-26 00:00:00 AMZN -100 @ 1003.73999"
[1] "2017-06-30 00:00:00 AMZN 100 @ 975.929993"
[1] "2017-07-03 00:00:00 AMZN -100 @ 968"
[1] "2017-07-07 00:00:00 AMZN 100 @ 965.140015"
[1] "2017-07-10 00:00:00 AMZN -100 @ 978.76001"
[1] "2017-07-11 00:00:00 AMZN 100 @ 996.469971"
[1] "2017-07-13 00:00:00 AMZN -100 @ 1006.51001"
[1] "2017-07-14 00:00:00 AMZN 100 @ 1000.630005"
[1] "2017-07-17 00:00:00 AMZN -100 @ 1001.809998"
[1] "2017-07-19 00:00:00 AMZN 100 @ 1024.449951"
[1] "2017-07-25 00:00:00 AMZN -100 @ 1038.949951"
[1] "2017-07-26 00:00:00 AMZN 100 @ 1039.869995"
[1] "2017-07-27 00:00:00 AMZN -100 @ 1052.800049"
[1] "2017-07-28 00:00:00 AMZN 100 @ 1046"
[1] "2017-07-31 00:00:00 AMZN -100 @ 1020.039978"
[1] "2017-08-03 00:00:00 AMZN 100 @ 995.890015"
[1] "2017-08-07 00:00:00 AMZN -100 @ 987.580017"
[1] "2017-08-08 00:00:00 AMZN 100 @ 992.27002"
[1] "2017-08-11 00:00:00 AMZN -100 @ 956.919983"
[1] "2017-08-15 00:00:00 AMZN 100 @ 983.299988"
[1] "2017-08-18 00:00:00 AMZN -100 @ 960.570007"
[1] "2017-08-24 00:00:00 AMZN 100 @ 958"
[1] "2017-08-25 00:00:00 AMZN -100 @ 952.450012"
[1] "2017-08-30 00:00:00 AMZN 100 @ 954.059998"
[1] "2017-09-06 00:00:00 AMZN -100 @ 965.27002"
[1] "2017-09-08 00:00:00 AMZN 100 @ 979.469971"
[1] "2017-09-12 00:00:00 AMZN -100 @ 977.960022"
[1] "2017-09-13 00:00:00 AMZN 100 @ 982.580017"
[1] "2017-09-18 00:00:00 AMZN -100 @ 986.789978"
[1] "2017-09-22 00:00:00 AMZN 100 @ 964.650024"
[1] "2017-09-25 00:00:00 AMZN -100 @ 955.099976"
[1] "2017-09-29 00:00:00 AMZN 100 @ 956.400024"
[1] "2017-10-06 00:00:00 AMZN 100 @ 980.849976"
[1] "2017-10-12 00:00:00 AMZN -200 @ 995"
[1] "2017-10-13 00:00:00 AMZN 100 @ 1000.929993"
[1] "2017-10-20 00:00:00 AMZN -100 @ 986.609985"
[1] "2017-10-26 00:00:00 AMZN 100 @ 972.909973"
[1] "2017-10-27 00:00:00 AMZN -100 @ 972.429993"
[1] "2017-10-31 00:00:00 AMZN 100 @ 1110.849976"
[1] "2017-11-02 00:00:00 AMZN -100 @ 1103.680054"
[1] "2017-11-07 00:00:00 AMZN 100 @ 1120.660034"
[1] "2017-11-09 00:00:00 AMZN -100 @ 1132.880005"
[1] "2017-11-10 00:00:00 AMZN 100 @ 1129.130005"
[1] "2017-11-13 00:00:00 AMZN -100 @ 1125.349976"
[1] "2017-11-16 00:00:00 AMZN 100 @ 1126.689941"
[1] "2017-11-17 00:00:00 AMZN -100 @ 1137.290039"
[1] "2017-11-20 00:00:00 AMZN 100 @ 1129.880005"
[1] "2017-11-21 00:00:00 AMZN -100 @ 1126.310059"
[1] "2017-11-24 00:00:00 AMZN 100 @ 1156.160034"
[1] "2017-11-30 00:00:00 AMZN -100 @ 1161.27002"
[1] "2017-12-04 00:00:00 AMZN 100 @ 1162.349976"
[1] "2017-12-05 00:00:00 AMZN -100 @ 1133.949951"
[1] "2017-12-07 00:00:00 AMZN 100 @ 1152.349976"
[1] "2017-12-13 00:00:00 AMZN 100 @ 1165.079956"
[1] "2017-12-14 00:00:00 AMZN -200 @ 1164.130005"
[1] "2017-12-18 00:00:00 AMZN 100 @ 1179.140015"
[1] "2017-12-21 00:00:00 AMZN -100 @ 1177.619995"
[1] "2017-12-28 00:00:00 AMZN 100 @ 1182.26001"
[1] "2018-01-03 00:00:00 AMZN -100 @ 1189.01001"
[1] "2018-01-04 00:00:00 AMZN 100 @ 1204.199951"
[1] "2018-01-09 00:00:00 AMZN 100 @ 1246.869995"
[1] "2018-01-12 00:00:00 AMZN -200 @ 1276.680054"
[1] "2018-01-16 00:00:00 AMZN 100 @ 1305.199951"
[1] "2018-01-18 00:00:00 AMZN -100 @ 1295"
[1] "2018-01-24 00:00:00 AMZN 100 @ 1362.540039"
[1] "2018-01-26 00:00:00 AMZN -100 @ 1377.949951"
[1] "2018-01-29 00:00:00 AMZN 100 @ 1402.050049"
[1] "2018-02-05 00:00:00 AMZN -100 @ 1429.949951"
[1] "2018-02-06 00:00:00 AMZN 100 @ 1390"
[1] "2018-02-07 00:00:00 AMZN -100 @ 1442.839966"
[1] "2018-02-08 00:00:00 AMZN 100 @ 1416.780029"
[1] "2018-02-09 00:00:00 AMZN -100 @ 1350.5"
[1] "2018-02-14 00:00:00 AMZN 100 @ 1414.51001"
[1] "2018-02-21 00:00:00 AMZN -100 @ 1468.349976"
[1] "2018-02-22 00:00:00 AMZN 100 @ 1482.920044"
[1] "2018-02-26 00:00:00 AMZN -100 @ 1500"
[1] "2018-02-27 00:00:00 AMZN 100 @ 1521.949951"
[1] "2018-03-01 00:00:00 AMZN -100 @ 1512.449951"
[1] "2018-03-06 00:00:00 AMZN 100 @ 1523.609985"
[1] "2018-03-13 00:00:00 AMZN 100 @ 1598.390015"
[1] "2018-03-15 00:00:00 AMZN -200 @ 1591"
[1] "2018-03-22 00:00:00 AMZN 100 @ 1581.859985"
[1] "2018-03-23 00:00:00 AMZN -100 @ 1544.920044"
[1] "2018-03-28 00:00:00 AMZN 100 @ 1497.050049"
[1] "2018-03-29 00:00:00 AMZN -100 @ 1431.420044"
[1] "2018-04-03 00:00:00 AMZN 100 @ 1371.98999"
[1] "2018-04-04 00:00:00 AMZN -100 @ 1392.050049"
[1] "2018-04-05 00:00:00 AMZN 100 @ 1410.569946"
[1] "2018-04-10 00:00:00 AMZN -100 @ 1406.079956"
[1] "2018-04-11 00:00:00 AMZN 100 @ 1436.219971"
[1] "2018-04-13 00:00:00 AMZN -100 @ 1448.5"
[1] "2018-04-16 00:00:00 AMZN 100 @ 1430.790039"
[1] "2018-04-17 00:00:00 AMZN -100 @ 1441.5"
[1] "2018-04-18 00:00:00 AMZN 100 @ 1503.829956"
[1] "2018-04-24 00:00:00 AMZN -100 @ 1517.859985"
[1] "2018-04-27 00:00:00 AMZN 100 @ 1517.959961"
[1] "2018-05-02 00:00:00 AMZN -100 @ 1582.26001"
[1] "2018-05-03 00:00:00 AMZN 100 @ 1569.680054"
[1] "2018-05-04 00:00:00 AMZN -100 @ 1572.079956"
[1] "2018-05-08 00:00:00 AMZN 100 @ 1600.140015"
[1] "2018-05-10 00:00:00 AMZN -100 @ 1608"
[1] "2018-05-11 00:00:00 AMZN 100 @ 1609.079956"
[1] "2018-05-14 00:00:00 AMZN -100 @ 1602.910034"
[1] "2018-05-18 00:00:00 AMZN 100 @ 1581.76001"
[1] "2018-05-21 00:00:00 AMZN -100 @ 1574.369995"
[1] "2018-05-23 00:00:00 AMZN 100 @ 1581.400024"
[1] "2018-05-24 00:00:00 AMZN -100 @ 1601.859985"
[1] "2018-05-25 00:00:00 AMZN 100 @ 1603.069946"
[1] "2018-05-30 00:00:00 AMZN 100 @ 1612.869995"
[1] "2018-06-01 00:00:00 AMZN 100 @ 1629.619995"
[1] "2018-06-05 00:00:00 AMZN 100 @ 1665.27002"
[1] "2018-06-08 00:00:00 AMZN -400 @ 1689.300049"
[1] "2018-06-14 00:00:00 AMZN 100 @ 1704.859985"
[1] "2018-06-18 00:00:00 AMZN 100 @ 1715.969971"
[1] "2018-06-19 00:00:00 AMZN -200 @ 1723.790039"
[1] "2018-06-21 00:00:00 AMZN 100 @ 1750.079956"
[1] "2018-06-25 00:00:00 AMZN -100 @ 1715.670044"
[1] "2018-06-28 00:00:00 AMZN 100 @ 1660.51001"
[1] "2018-06-29 00:00:00 AMZN -100 @ 1701.449951"
[1] "2018-07-02 00:00:00 AMZN 100 @ 1699.800049"
[1] "2018-07-03 00:00:00 AMZN -100 @ 1713.780029"
[1] "2018-07-05 00:00:00 AMZN 100 @ 1693.959961"
[1] "2018-07-06 00:00:00 AMZN -100 @ 1699.72998"
[1] "2018-07-09 00:00:00 AMZN 100 @ 1710.630005"
[1] "2018-07-13 00:00:00 AMZN 100 @ 1796.619995"
[1] "2018-07-19 00:00:00 AMZN 100 @ 1842.920044"
[1] "2018-07-20 00:00:00 AMZN -300 @ 1812.969971"
[1] "2018-07-26 00:00:00 AMZN 100 @ 1863.609985"
[1] "2018-07-30 00:00:00 AMZN -100 @ 1817.27002"
[1] "2018-07-31 00:00:00 AMZN 100 @ 1779.219971"
[1] "2018-08-01 00:00:00 AMZN -100 @ 1777.439941"
[1] "2018-08-03 00:00:00 AMZN 100 @ 1834.329956"
[1] "2018-08-07 00:00:00 AMZN -100 @ 1847.75"
[1] "2018-08-08 00:00:00 AMZN 100 @ 1862.47998"
[1] "2018-08-14 00:00:00 AMZN -100 @ 1896.199951"
[1] "2018-08-15 00:00:00 AMZN 100 @ 1919.650024"
[1] "2018-08-17 00:00:00 AMZN -100 @ 1886.52002"
[1] "2018-08-23 00:00:00 AMZN 100 @ 1904.900024"
[1] "2018-08-27 00:00:00 AMZN -100 @ 1905.390015"
[1] "2018-08-29 00:00:00 AMZN 100 @ 1932.819946"
[1] "2018-08-31 00:00:00 AMZN 100 @ 2002.380005"
[1] "2018-09-04 00:00:00 AMZN -200 @ 2012.709961"
[1] "2018-09-06 00:00:00 AMZN 100 @ 1994.819946"
[1] "2018-09-07 00:00:00 AMZN -100 @ 1958.310059"
[1] "2018-09-13 00:00:00 AMZN 100 @ 1990"
[1] "2018-09-17 00:00:00 AMZN -100 @ 1970.189941"
[1] "2018-09-20 00:00:00 AMZN 100 @ 1926.420044"
[1] "2018-09-21 00:00:00 AMZN -100 @ 1944.300049"
[1] "2018-09-24 00:00:00 AMZN 100 @ 1915.01001"
[1] "2018-09-25 00:00:00 AMZN -100 @ 1934.359985"
[1] "2018-09-26 00:00:00 AMZN 100 @ 1974.550049"
[1] "2018-09-28 00:00:00 AMZN -100 @ 2012.97998"
[1] "2018-10-01 00:00:00 AMZN 100 @ 2003"
[1] "2018-10-02 00:00:00 AMZN -100 @ 2004.359985"
[1] "2018-10-11 00:00:00 AMZN 100 @ 1755.25"
[1] "2018-10-12 00:00:00 AMZN -100 @ 1719.359985"
[1] "2018-10-16 00:00:00 AMZN 100 @ 1760.949951"
[1] "2018-10-17 00:00:00 AMZN -100 @ 1819.959961"
[1] "2018-10-18 00:00:00 AMZN 100 @ 1831.72998"
[1] "2018-10-22 00:00:00 AMZN -100 @ 1764.030029"
[1] "2018-10-23 00:00:00 AMZN 100 @ 1789.300049"
[1] "2018-10-25 00:00:00 AMZN -100 @ 1664.199951"
[1] "2018-10-29 00:00:00 AMZN 100 @ 1642.810059"
[1] "2018-10-30 00:00:00 AMZN -100 @ 1538.880005"
[1] "2018-11-01 00:00:00 AMZN 100 @ 1598.01001"
[1] "2018-11-07 00:00:00 AMZN -100 @ 1642.810059"
[1] "2018-11-08 00:00:00 AMZN 100 @ 1755.48999"
[1] "2018-11-12 00:00:00 AMZN -100 @ 1712.430054"
[1] "2018-11-19 00:00:00 AMZN 100 @ 1593.410034"
[1] "2018-11-20 00:00:00 AMZN -100 @ 1512.290039"
[1] "2018-11-26 00:00:00 AMZN 100 @ 1502.060059"
[1] "2018-11-27 00:00:00 AMZN -100 @ 1581.329956"
[1] "2018-11-28 00:00:00 AMZN 100 @ 1581.420044"
[1] "2018-12-03 00:00:00 AMZN -100 @ 1690.170044"
[1] "2018-12-04 00:00:00 AMZN 100 @ 1772.359985"
[1] "2018-12-07 00:00:00 AMZN -100 @ 1699.189941"
[1] "2018-12-10 00:00:00 AMZN 100 @ 1629.130005"
[1] "2018-12-11 00:00:00 AMZN -100 @ 1641.030029"
[1] "2018-12-12 00:00:00 AMZN 100 @ 1643.23999"
[1] "2018-12-17 00:00:00 AMZN -100 @ 1591.910034"
[1] "2018-12-20 00:00:00 AMZN 100 @ 1495.079956"
[1] "2018-12-21 00:00:00 AMZN -100 @ 1460.829956"
[1] "2018-12-28 00:00:00 AMZN 100 @ 1461.640015"
[1] "2019-01-02 00:00:00 AMZN 100 @ 1501.969971"
[1] "2019-01-07 00:00:00 AMZN -200 @ 1575.390015"
[1] "2019-01-08 00:00:00 AMZN 100 @ 1629.51001"
[1] "2019-01-11 00:00:00 AMZN -100 @ 1656.219971"
[1] "2019-01-17 00:00:00 AMZN 100 @ 1683.780029"
[1] "2019-01-23 00:00:00 AMZN -100 @ 1632.170044"
[1] "2019-01-25 00:00:00 AMZN 100 @ 1654.930054"
[1] "2019-01-30 00:00:00 AMZN -100 @ 1593.880005"
[1] "2019-02-01 00:00:00 AMZN 100 @ 1718.72998"
[1] "2019-02-05 00:00:00 AMZN -100 @ 1633.310059"
[1] "2019-02-06 00:00:00 AMZN 100 @ 1658.810059"
[1] "2019-02-08 00:00:00 AMZN -100 @ 1614.369995"
[1] "2019-02-13 00:00:00 AMZN 100 @ 1638.01001"
[1] "2019-02-19 00:00:00 AMZN -100 @ 1607.949951"
[1] "2019-02-21 00:00:00 AMZN 100 @ 1622.099976"
[1] "2019-02-22 00:00:00 AMZN -100 @ 1619.439941"
[1] "2019-02-26 00:00:00 AMZN 100 @ 1633"
[1] "2019-02-28 00:00:00 AMZN 100 @ 1641.089966"
[1] "2019-03-04 00:00:00 AMZN -200 @ 1671.72998"
[1] "2019-03-05 00:00:00 AMZN 100 @ 1696.170044"
[1] "2019-03-07 00:00:00 AMZN -100 @ 1668.949951"
[1] "2019-03-13 00:00:00 AMZN 100 @ 1673.099976"
[1] "2019-03-15 00:00:00 AMZN 100 @ 1686.219971"
[1] "2019-03-18 00:00:00 AMZN -200 @ 1712.359985"
[1] "2019-03-19 00:00:00 AMZN 100 @ 1742.150024"
[1] "2019-03-26 00:00:00 AMZN -100 @ 1774.26001"
[1] "2019-03-27 00:00:00 AMZN 100 @ 1783.76001"
[1] "2019-03-29 00:00:00 AMZN -100 @ 1773.420044"
[1] "2019-04-01 00:00:00 AMZN 100 @ 1780.75"
[1] "2019-04-04 00:00:00 AMZN -100 @ 1820.699951"
[1] "2019-04-09 00:00:00 AMZN 100 @ 1849.859985"
[1] "2019-04-11 00:00:00 AMZN -100 @ 1847.329956"
[1] "2019-04-12 00:00:00 AMZN 100 @ 1844.069946"
[1] "2019-04-15 00:00:00 AMZN -100 @ 1843.060059"
[1] "2019-04-18 00:00:00 AMZN 100 @ 1864.819946"
[1] "2019-04-22 00:00:00 AMZN -100 @ 1861.689941"
[1] "2019-04-24 00:00:00 AMZN 100 @ 1923.77002"
[1] "2019-04-26 00:00:00 AMZN -100 @ 1902.25"
[1] "2019-04-30 00:00:00 AMZN 100 @ 1938.430054"
[1] "2019-05-01 00:00:00 AMZN -100 @ 1926.52002"
[1] "2019-05-07 00:00:00 AMZN 100 @ 1950.550049"
[1] "2019-05-08 00:00:00 AMZN -100 @ 1921"
[1] "2019-05-16 00:00:00 AMZN 100 @ 1871.150024"
[1] "2019-05-21 00:00:00 AMZN -100 @ 1858.969971"
[1] "2019-05-24 00:00:00 AMZN 100 @ 1815.47998"
[1] "2019-05-28 00:00:00 AMZN -100 @ 1823.280029"
[1] "2019-05-29 00:00:00 AMZN 100 @ 1836.430054"
[1] "2019-05-31 00:00:00 AMZN -100 @ 1816.319946"
[1] "2019-06-06 00:00:00 AMZN 100 @ 1738.5"
[1] "2019-06-14 00:00:00 AMZN -100 @ 1870.300049"
[1] "2019-06-17 00:00:00 AMZN 100 @ 1869.670044"
[1] "2019-06-18 00:00:00 AMZN -100 @ 1886.030029"
[1] "2019-06-19 00:00:00 AMZN 100 @ 1901.369995"
[1] "2019-06-24 00:00:00 AMZN 100 @ 1911.300049"
[1] "2019-06-25 00:00:00 AMZN -200 @ 1913.900024"
[1] "2019-06-28 00:00:00 AMZN 100 @ 1904.280029"
[1] "2019-07-02 00:00:00 AMZN -100 @ 1922.189941"
[1] "2019-07-03 00:00:00 AMZN 100 @ 1934.310059"
[1] "2019-07-10 00:00:00 AMZN 100 @ 1988.300049"
[1] "2019-07-15 00:00:00 AMZN -200 @ 2011"
[1] "2019-07-24 00:00:00 AMZN 100 @ 1994.48999"
[1] "2019-07-29 00:00:00 AMZN -100 @ 1943.050049"
[1] "2019-08-08 00:00:00 AMZN 100 @ 1793.400024"
[1] "2019-08-13 00:00:00 AMZN -100 @ 1784.920044"
[1] "2019-08-15 00:00:00 AMZN 100 @ 1762.959961"
[1] "2019-08-16 00:00:00 AMZN -100 @ 1776.119995"
[1] "2019-08-19 00:00:00 AMZN 100 @ 1792.569946"
[1] "2019-08-22 00:00:00 AMZN -100 @ 1823.540039"
[1] "2019-08-23 00:00:00 AMZN 100 @ 1804.660034"
[1] "2019-08-26 00:00:00 AMZN -100 @ 1749.619995"
[1] "2019-08-28 00:00:00 AMZN 100 @ 1761.829956"
[1] "2019-08-30 00:00:00 AMZN 100 @ 1786.400024"
[1] "2019-09-04 00:00:00 AMZN -200 @ 1789.839966"
[1] "2019-09-05 00:00:00 AMZN 100 @ 1800.619995"
[1] "2019-09-10 00:00:00 AMZN -100 @ 1831.349976"
[1] "2019-09-16 00:00:00 AMZN 100 @ 1839.339966"
[1] "2019-09-17 00:00:00 AMZN -100 @ 1807.839966"
[1] "2019-09-19 00:00:00 AMZN 100 @ 1817.459961"
[1] "2019-09-20 00:00:00 AMZN -100 @ 1821.5"
[1] "2019-09-23 00:00:00 AMZN 100 @ 1794.160034"
[1] "2019-09-24 00:00:00 AMZN -100 @ 1785.300049"
[1] "2019-09-27 00:00:00 AMZN 100 @ 1739.839966"
[1] "2019-09-30 00:00:00 AMZN -100 @ 1725.449951"
[1] "2019-10-02 00:00:00 AMZN 100 @ 1735.650024"
[1] "2019-10-04 00:00:00 AMZN -100 @ 1724.420044"
[1] "2019-10-07 00:00:00 AMZN 100 @ 1739.650024"
[1] "2019-10-09 00:00:00 AMZN -100 @ 1705.51001"
[1] "2019-10-11 00:00:00 AMZN 100 @ 1720.26001"
[1] "2019-10-15 00:00:00 AMZN 100 @ 1736.430054"
[1] "2019-10-22 00:00:00 AMZN -200 @ 1785.660034"
[1] "2019-10-23 00:00:00 AMZN 100 @ 1765.72998"
[1] "2019-10-24 00:00:00 AMZN -100 @ 1762.170044"
[1] "2019-10-28 00:00:00 AMZN 100 @ 1761.329956"
[1] "2019-10-29 00:00:00 AMZN -100 @ 1777.079956"
[1] "2019-10-30 00:00:00 AMZN 100 @ 1762.709961"
[1] "2019-10-31 00:00:00 AMZN -100 @ 1779.98999"
[1] "2019-11-01 00:00:00 AMZN 100 @ 1776.660034"
[1] "2019-11-04 00:00:00 AMZN -100 @ 1791.439941"
[1] "2019-11-05 00:00:00 AMZN 100 @ 1804.660034"
[1] "2019-11-07 00:00:00 AMZN -100 @ 1795.77002"
[1] "2019-11-14 00:00:00 AMZN 100 @ 1753.109985"
[1] "2019-11-15 00:00:00 AMZN -100 @ 1754.599976"
[1] "2019-11-18 00:00:00 AMZN 100 @ 1739.48999"
[1] "2019-11-19 00:00:00 AMZN -100 @ 1752.530029"
[1] "2019-11-20 00:00:00 AMZN 100 @ 1752.790039"
[1] "2019-11-22 00:00:00 AMZN -100 @ 1734.709961"
[1] "2019-11-26 00:00:00 AMZN 100 @ 1773.839966"
[1] "2019-12-03 00:00:00 AMZN -100 @ 1781.599976"
[1] "2019-12-10 00:00:00 AMZN 100 @ 1749.51001"
[1] "2019-12-12 00:00:00 AMZN -100 @ 1748.719971"
[1] "2019-12-13 00:00:00 AMZN 100 @ 1760.329956"
[1] "2019-12-18 00:00:00 AMZN 100 @ 1790.660034"
[1] "2019-12-20 00:00:00 AMZN -200 @ 1792.280029"
[1] "2019-12-23 00:00:00 AMZN 100 @ 1786.5"
[1] "2019-12-24 00:00:00 AMZN -100 @ 1793"
[1] "2019-12-26 00:00:00 AMZN 100 @ 1789.209961"
[1] "2019-12-27 00:00:00 AMZN -100 @ 1868.77002"
[1] "2019-12-30 00:00:00 AMZN 100 @ 1869.800049"
[1] "2019-12-31 00:00:00 AMZN -100 @ 1846.890015"
_images/r_strategies_12_1.png 
[12]:

tstats <- tradeStats(portfolio.st, use="trades", inclZeroDays=FALSE)
data.frame(t(tstats))
A data.frame: 32 × 1
AMZN
<chr>
Portfoliomy strategy
SymbolAMZN
Num.Txns348
Num.Trades160
Net.Trading.PL103829
Avg.Trade.PL648.9312
Med.Trade.PL194.4946
Largest.Winner24637.02
Largest.Loser-12510.01
Gross.Profits297592
Gross.Losses-193763
Std.Dev.Trade.PL4650.819
Std.Err.Trade.PL367.6795
Percent.Positive53.125
Percent.Negative46.875
Profit.Factor1.535856
Avg.Win.Trade3501.083
Med.Win.Trade1707.996
Avg.Losing.Trade-2583.507
Med.Losing.Trade-1808.008
Avg.Daily.PL648.9312
Med.Daily.PL194.4947
Std.Dev.Daily.PL4650.819
Std.Err.Daily.PL367.6795
Ann.Sharpe2.214979
Max.Drawdown-39630.02
Profit.To.Max.Draw2.619958
Avg.WinLoss.Ratio1.355167
Med.WinLoss.Ratio0.9446838
Max.Equity106120
Min.Equity-2470.996
End.Equity103829 
[13]:

#install.packages("magrittr") # package installations are only needed the first time you use it
#install.packages("dplyr")    # alternative installation of the %>%
library(magrittr) # needs to be run every time you start R and want to use %>%
library(dplyr)    # alternatively, this also loads %>%
trades <- tstats %>%
    mutate(Trades = Num.Trades,
        Win.Percent = Percent.Positive,
        Loss.Percent = Percent.Negative,
        WL.Ratio = Percent.Positive/Percent.Negative) %>%
    select(Trades, Win.Percent, Loss.Percent, WL.Ratio)
data.frame(t(trades))


Attaching package: ‘dplyr’


The following objects are masked from ‘package:xts’:

    first, last


The following objects are masked from ‘package:stats’:

    filter, lag


The following objects are masked from ‘package:base’:

    intersect, setdiff, setequal, union
A data.frame: 4 × 1
t.trades.
<dbl>
Trades160.000000
Win.Percent 53.125000
Loss.Percent 46.875000
WL.Ratio 1.133333
 
[14]:

a <- getAccount(account.st)
equity <- a$summary$End.Eq
plot(equity, main = "Equity Curve QQQ")
_images/r_strategies_15_0.png 
[15]:

portfolio <- getPortfolio(portfolio.st)
portfolioSummary <- portfolio$summary
colnames(portfolio$summary)
tail(portfolio$summary)
  1. 'Long.Value'
  2. 'Short.Value'
  3. 'Net.Value'
  4. 'Gross.Value'
  5. 'Period.Realized.PL'
  6. 'Period.Unrealized.PL'
  7. 'Gross.Trading.PL'
  8. 'Txn.Fees'
  9. 'Net.Trading.PL'
           Long.Value Short.Value Net.Value Gross.Value Period.Realized.PL
2019-12-23     178650           0    178650      178650              0.000
2019-12-24          0           0         0           0            650.000
2019-12-26     178921           0    178921      178921              0.000
2019-12-27          0           0         0           0           7956.006
2019-12-30     186980           0    186980      186980              0.000
2019-12-31          0           0         0           0          -2291.003
           Period.Unrealized.PL Gross.Trading.PL Txn.Fees Net.Trading.PL
2019-12-23                    0            0.000        0          0.000
2019-12-24                    0          650.000        0        650.000
2019-12-26                    0            0.000        0          0.000
2019-12-27                    0         7956.006        0       7956.006
2019-12-30                    0            0.000        0          0.000
2019-12-31                    0        -2291.003        0      -2291.003

[16]:

ret <- Return.calculate(equity, method = "log")
tail(ret)
charts.PerformanceSummary(ret, colorset = bluefocus, main = "Strategy Performance")

                 End.Eq
2019-12-23  0.000000000
2019-12-24  0.003285503
2019-12-26  0.000000000
2019-12-27  0.039363582
2019-12-30  0.000000000
2019-12-31 -0.011177133
_images/r_strategies_17_1.png 
[17]:

rets <- PortfReturns(Account = account.st)
chart.Boxplot(rets, main = "QQQ Returns", colorset= rich10equal)
_images/r_strategies_18_0.png 
[18]:

table.Drawdowns(rets, top=10)
A data.frame: 10 × 7
FromTroughToDepthLengthTo TroughRecovery
<date><date><date><dbl><dbl><dbl><dbl>
2018-07-192019-02-222019-12-27-0.3726364150214
2018-03-142018-03-292018-05-10-0.1135 41 12 29
2017-07-312017-09-112018-01-12-0.0966116 30 86
2017-04-072017-04-202017-05-23-0.0671 32 9 23
2018-02-092018-02-092018-02-26-0.0663 11 1 10
2018-02-022018-02-022018-02-07-0.0609 4 1 3
2017-06-012017-06-212017-07-27-0.0535 40 15 25
2018-06-072018-06-252018-06-29-0.0362 17 13 4
2017-02-062017-02-072017-02-21-0.0322 11 2 9
2019-12-312019-12-31NA-0.0229 2 1 NA
 
[19]:

table.CalendarReturns(rets)
A data.frame: 3 × 13
JanFebMarAprMayJunJulAugSepOctNovDecAMZN.DailyEqPL
<dbl><dbl><dbl><dbl><dbl><dbl><dbl><dbl><dbl><dbl><dbl><dbl><dbl>
20170 0.0 0.40.9 0.30.0-2.61.4 0.00.0-3.2 0.4-2.6
20182-1.0-6.65.5 2.44.1 0.00.4 3.80.0-0.4 1.612.0
20190 0.5 0.80.0-0.30.0 0.02.2-1.41.7 2.2-2.3 3.3
 
[20]:

table.DownsideRisk(rets)
A data.frame: 11 × 1
AMZN.DailyEqPL
<dbl>
Semi Deviation 0.0143
Gain Deviation 0.0208
Loss Deviation 0.0225
Downside Deviation (MAR=210%) 0.0177
Downside Deviation (Rf=0%) 0.0138
Downside Deviation (0%) 0.0138
Maximum Drawdown 0.3726
Historical VaR (95%)-0.0277
Historical ES (95%)-0.0528
Modified VaR (95%)-0.0264
Modified ES (95%)-0.0264
 
[21]:

table.Stats(rets)
A data.frame: 16 × 1
AMZN.DailyEqPL
<dbl>
Observations753.0000
NAs 0.0000
Minimum -0.1045
Quartile 1 0.0000
Median 0.0000
Arithmetic Mean 0.0014
Geometric Mean 0.0012
Quartile 3 0.0029
Maximum 0.1502
SE Mean 0.0007
LCL Mean (0.95) -0.0001
UCL Mean (0.95) 0.0028
Variance 0.0004
Stdev 0.0206
Skewness 0.2160
Kurtosis 11.3408
 
[22]:

chart.RiskReturnScatter(rets)
_images/r_strategies_23_0.png 
[23]:

chart.ME(Portfolio = portfolio.st, Symbol = symbol, type = "MAE", scale = "percent")
_images/r_strategies_24_0.png 
[24]:

chart.ME(Portfolio = portfolio.st, Symbol = symbol, type = "MFE", scale = "percent")
_images/r_strategies_25_0.png
Moving Average Crossover Strategy - Sample 2

  • when the short term moving average crosses above the long term moving average, this indicates a buy signal

  • when the short term moving average crosses below the long term moving average, it may be a good moment to sell

Reference: https://towardsdatascience.com/making-a-trade-call-using-simple-moving-average-sma-crossover-strategy-python-implementation-29963326da7a

[25]:

# Moving Average Crossover Strategy - Sample 2
# strategy resetting
rm.strat("my strategy")
out <- initPortf(portfolio.st, symbols = symbol, initDate = start, currency = "USD")
out <- initAcct(account.st, portfolios = portfolio.st, initDate = start, currency = "USD", initEq = initEquity)
initOrders(portfolio.st, initDate = start)
strategy(strategy.st, store = TRUE)
# Indicators
out <- add.indicator(strategy.st, name = "SMA", arguments = list(x = quote(Cl(AMZN)), n = 20, maType = "SMA"), label = "SMA20periods")
out <- add.indicator(strategy.st, name = "SMA", arguments = list(x = quote(Cl(AMZN)), n = 50, maType = "SMA"), label = "SMA50periods")
out <- add.indicator(strategy.st, name = "EMA", arguments = list(x = quote(Cl(AMZN)), n = 20, maType = "EMA"), label = "EMA20periods")
out <- add.indicator(strategy.st, name = "EMA", arguments = list(x = quote(Cl(AMZN)), n = 50, maType = "EMA"), label = "EMA50periods")
# Signals
#out <- add.signal(strategy.st, name = "sigCrossover", arguments = list(columns = c("SMA20periods", "SMA50periods"), relationship = "gte", cross = TRUE), label = "buy")
#out <- add.signal(strategy.st, name = "sigCrossover", arguments = list(columns = c("SMA20periods", "SMA50periods"), relationship = "lte", cross = TRUE), label = "sell")
out <- add.signal(strategy.st, name = "sigCrossover", arguments = list(columns = c("EMA20periods", "EMA50periods"), relationship = "gte", cross = TRUE), label = "buy2")
out <- add.signal(strategy.st, name = "sigCrossover", arguments = list(columns = c("EMA20periods", "EMA50periods"), relationship = "lte", cross = TRUE), label = "sell2")
# Rules
out <- add.rule(strategy.st, name = 'ruleSignal', arguments = list(sigcol = "buy2", sigval = TRUE, orderqty = 100, ordertype = 'market', orderside = 'long'), type = 'enter')
out <- add.rule(strategy.st, name = 'ruleSignal', arguments = list(sigcol = "sell2", sigval = TRUE, orderqty = 'all', ordertype = 'market', orderside = 'long'), type = 'exit')
#out <- add.rule(strategy.st, name = 'ruleSignal', arguments = list(sigcol = "sell2", sigval = TRUE, orderqty = -100, ordertype = 'market', orderside = 'short'), type = 'enter')
#out <- add.rule(strategy.st, name = 'ruleSignal', arguments = list(sigcol = "buy2", sigval = TRUE, orderqty = 'all', ordertype = 'market', orderside = 'short'), type = 'exit')

[26]:

# backtesting
applyStrategy(strategy.st, portfolios=portfolio.st)
out <- updatePortf(portfolio.st)
dateRange <- time(getPortfolio(portfolio.st)$summary)[-1]
out <- updateAcct(portfolio.st, dateRange)
out <- updateEndEq(account.st)
chart.Posn(portfolio.st, Symbol=symbol) #, TA=c("add_SMA(n=20,col='red')"))

[1] "2017-10-16 00:00:00 AMZN 100 @ 1002.940002"
[1] "2018-10-16 00:00:00 AMZN -100 @ 1760.949951"
[1] "2019-02-01 00:00:00 AMZN 100 @ 1718.72998"
[1] "2019-02-13 00:00:00 AMZN -100 @ 1638.01001"
[1] "2019-03-04 00:00:00 AMZN 100 @ 1671.72998"
[1] "2019-06-06 00:00:00 AMZN -100 @ 1738.5"
[1] "2019-06-14 00:00:00 AMZN 100 @ 1870.300049"
[1] "2019-08-08 00:00:00 AMZN -100 @ 1793.400024"
[1] "2019-12-27 00:00:00 AMZN 100 @ 1868.77002"
_images/r_strategies_28_1.png 
[27]:

tstats <- tradeStats(portfolio.st, use="trades", inclZeroDays=FALSE)
data.frame(t(tstats))
A data.frame: 32 × 1
AMZN
<chr>
Portfoliomy strategy
SymbolAMZN
Num.Txns9
Num.Trades5
Net.Trading.PL64528
Avg.Trade.PL12905.6
Med.Trade.PL-2188
Largest.Winner75800.99
Largest.Loser-8071.997
Gross.Profits82478
Gross.Losses-17950
Std.Dev.Trade.PL35660.49
Std.Err.Trade.PL15947.85
Percent.Positive40
Percent.Negative60
Profit.Factor4.594874
Avg.Win.Trade41239
Med.Win.Trade41239
Avg.Losing.Trade-5983.333
Med.Losing.Trade-7690.003
Avg.Daily.PL16679
Med.Daily.PL-506.5003
Std.Dev.Daily.PL40007.96
Std.Err.Daily.PL20003.98
Ann.Sharpe6.617956
Max.Drawdown-41021
Profit.To.Max.Draw1.573048
Avg.WinLoss.Ratio6.892312
Med.WinLoss.Ratio5.362677
Max.Equity103657
Min.Equity-3664.001
End.Equity64528 
[28]:

#install.packages("magrittr") # package installations are only needed the first time you use it
#install.packages("dplyr")    # alternative installation of the %>%
#library(magrittr) # needs to be run every time you start R and want to use %>%
#library(dplyr)    # alternatively, this also loads %>%
trades <- tstats %>%
    mutate(Trades = Num.Trades,
        Win.Percent = Percent.Positive,
        Loss.Percent = Percent.Negative,
        WL.Ratio = Percent.Positive/Percent.Negative) %>%
    select(Trades, Win.Percent, Loss.Percent, WL.Ratio)
data.frame(t(trades))
A data.frame: 4 × 1
t.trades.
<dbl>
Trades 5.0000000
Win.Percent40.0000000
Loss.Percent60.0000000
WL.Ratio 0.6666667
 
[29]:

a <- getAccount(account.st)
equity <- a$summary$End.Eq
plot(equity, main = "Equity Curve QQQ")
_images/r_strategies_31_0.png 
[30]:

ret <- Return.calculate(equity, method = "log")
charts.PerformanceSummary(ret, colorset = bluefocus, main = "Strategy Performance")
_images/r_strategies_32_0.png 
[31]:

rets <- PortfReturns(Account = account.st)
chart.RiskReturnScatter(rets)
_images/r_strategies_33_0.png

Python

Prerequisites

For using the program language Python, you don’t have to install nothing: Python is already installed.

There are a lot of guide for getting started with Python: knowing the basics will be taken for granted.

But you have to install some specific packages and it is a best practice to use a virtual environment. For unix,

$ git clone https://github.com/bilardi/backtesting-tool-comparison
$ cd backtesting-tool-comparison/
$ python3 -m venv .env # create virtual environment
$ source .env/bin/activate # enter in the virtual environment
$ pip install --upgrade -r requirements.txt # install your dependences

And when you want to delete all environment,

$ deactivate # exit when you will have finished the job
$ rm -rf .env # remove the virtual environment it is a best practice

A short description of each package that you have to import is below.

$ python
import yfinance # for Yahoo Finance historical data of stock market
import yahoo-fin # for Yahoo Finance historical data of stock market
import yahoofinancials # for Yahoo Finance historical data of stock market
import pandas_datareader # for historical data many sources
import datetime # for datetime management
import requests # for getting contents by url
import quandl # for Quandl historical data
import matplotlib # for plots
import mplfinance # for financial plots
import pandas # for data management
import numpy # for data management
import backtesting # for strategy management

If you want to use Jupyter, you can use directly the commands below

$ git clone https://github.com/bilardi/backtesting-tool-comparison
$ cd backtesting-tool-comparison/
$ pip install --upgrade -r requirements.txt
$ docker run --rm -p 8888:8888 -e JUPYTER_ENABLE_LAB=yes -v "$PWD":/home/jovyan/ jupyter/scipy-notebook

Jupyter is very easy for having a GUI virtual environment: knowing the basics will be taken for granted.

You can find all scripts descripted in this tutorial on GitHub.

  • the .py files in src/python/ folder, the you can use on shell

  • the .ipynb files in docs/sources/python/ folder, that you can use on Jupyter or browse on this tutorial

Getting started

Python is a programming language and it has more libraries dedicated for statistical analysis, graphics representation and reporting.

This is a summary of the Python language syntax: you can find more details in python documentation.

Basics

[1]:

# initialization of a variable
my_numeric_variable = 1
my_string_variable = "hello"

[2]:

# print a variable directly
my_numeric_variable

[2]:

1

[3]:

# print method
print(my_string_variable)

hello

Vectors

[4]:

# initialization of a vector
my_numeric_vector = [1,2,3,4,5,6]
my_sequence = list(range(1,7))
my_string_vector = ["hello", "world", "!"]
my_logic_vector = [True, False]
import random
my_random_vector = random.sample(list(range(1,5000)), 25) # Vector with 25 elements with random number from 1 to 5000

[5]:

my_numeric_vector

[5]:

[1, 2, 3, 4, 5, 6]
Operations with vectors

There are many libraries for statistics analysis, so you can use the same operations on different libraries.

[6]:

sum(my_numeric_vector) # sum
import statistics
print(statistics.mean(my_numeric_vector)) # mean
print(statistics.median(my_numeric_vector)) # median
import numpy
print(numpy.median(my_numeric_vector)) # median

3.5
3.5
3.5

[7]:

# multiplication 2 with each element of a vector
# each variable my_new_vector is the same
my_new_vector = [e * 2 for e in my_numeric_vector]
my_new_vector = list(map(lambda x: x * 2, my_numeric_vector))
import pandas
serie = pandas.Series(my_numeric_vector)
my_new_vector = (serie * 2).tolist()
import numpy
my_new_vector = list(numpy.array(my_numeric_vector) * 2)
my_new_vector

[7]:

[2, 4, 6, 8, 10, 12]

[8]:

# division 2 with each element of a vector
# each variable my_new_vector is the same
my_new_vector = [e / 2 for e in my_numeric_vector]
my_new_vector = list(map(lambda x: x / 2, my_numeric_vector))
serie = pandas.Series(my_numeric_vector)
my_new_vector = (serie / 2).tolist()
my_new_vector = list(numpy.array(my_numeric_vector) / 2)
my_new_vector

[8]:

[0.5, 1.0, 1.5, 2.0, 2.5, 3.0]

[9]:

# sum each element of a vector with another vector by position
# each variable my_new_vector is the same
my_new_vector = [sum(e) for e in zip(my_numeric_vector, my_sequence)]
my_new_vector = [x + y for x, y in zip(my_numeric_vector, my_sequence)]
import operator
my_new_vector = list(map(operator.add, my_numeric_vector, my_sequence))
import pandas
my_new_vector = list(pandas.Series(my_numeric_vector).add(pandas.Series(my_sequence)))
import numpy
my_new_vector = list(numpy.array(my_numeric_vector) + numpy.array(my_sequence))
my_new_vector

[9]:

[2, 4, 6, 8, 10, 12]

[10]:

# get element from a vector
print(my_string_vector[0]) # print hello
print(my_string_vector[::len(my_string_vector)-1]) # print hello!
print(my_string_vector[0:2]) # print hello world

hello
['hello', '!']
['hello', 'world']

[11]:

# add labels to a vector
import pandas
names = ["one","two","three","four","five","six"]
my_vector = pandas.DataFrame(my_numeric_vector, index=names).T
print(my_vector)

   one  two  three  four  five  six
0    1    2      3     4     5    6

[12]:

# get data type of a vector
import numpy
print(numpy.array(my_vector).dtype.type) # print numpy.int64
print(numpy.array(my_string_vector).dtype.type) # print numpy.str_

<class 'numpy.int64'>
<class 'numpy.str_'>

[13]:

# convertion of each element of a vector
# each variable my_new_string_vector is the same
my_new_string_vector = [str(e) for e in my_numeric_vector]
my_new_string_vector = list(map(str, my_numeric_vector))
my_new_string_vector

[13]:

['1', '2', '3', '4', '5', '6']

Matrixes

A matrix is a vector with more dimensions

[14]:

my_matrix = [[0] * 5 for e in range(2)] # creates a matrix bidimensional with 2 rows and 5 columns
my_matrix

[14]:

[[0, 0, 0, 0, 0], [0, 0, 0, 0, 0]]

[15]:

my_matrix = [list(range(1,6)) for e in range(2)] # creates a matrix bidimensional with 2 rows and 5 columns
my_matrix

[15]:

[[1, 2, 3, 4, 5], [1, 2, 3, 4, 5]]

[16]:

import numpy
my_matrix = numpy.arange(1,11).reshape(2, 5) # creates a matrix bidimensional with 2 rows and 5 columns
my_matrix

[16]:

array([[ 1,  2,  3,  4,  5],
       [ 6,  7,  8,  9, 10]])

[17]:

my_matrix = numpy.arange(1,11).reshape(5, 2).T # creates a matrix bidimensional with 2 rows and 5 columns like R language
my_matrix

[17]:

array([[ 1,  3,  5,  7,  9],
       [ 2,  4,  6,  8, 10]])

[18]:

print(my_matrix[1,1]) # prints 4
print(my_matrix[0,]) # prints row 1
print(my_matrix[:,1]) # prints column 2

4
[1 3 5 7 9]
[3 4]

[19]:

# merge of vectors
vector_one = ["one", 0.1]
vector_two = ["two", 1]
vector_three = ["three", 10]
my_vectors = [vector_one, vector_two, vector_three]
import pandas
colnames = ["vector number", "quantity"]
rownames = ["yesterday", "today", "tomorrow"]
pandas.DataFrame(my_vectors, index=rownames, columns=colnames)

[19]:
vector number quantity
yesterday one 0.1
today two 1.0
tomorrow three 10.0

Weighted average

[20]:

# performance
apple_performance = 3
netflix_performance = 7
amazon_performance = 11
# weight
apple_weight = .3
netflix_weight = .4
amazon_weight = .3

[21]:

# portfolio performance
weighted_average = apple_performance * apple_weight + netflix_performance * netflix_weight + amazon_performance * amazon_weight
weighted_average

[21]:

7.0

[22]:

# the same sample but with vectors
performance = [3,7,11]
weight = [.3,.4,.3]
company = ['apple','netflix','amazon']
import pandas
performance = pandas.DataFrame(performance, index=company).T
weight = pandas.DataFrame(weight, index=company).T
print(performance)
print(weight)

   apple  netflix  amazon
0      3        7      11
   apple  netflix  amazon
0    0.3      0.4     0.3

[23]:

# with headers
performance_weight = performance.multiply(weight)
print(performance_weight)
weighted_average = performance_weight.sum(axis=1)
print(weighted_average)

   apple  netflix  amazon
0    0.9      2.8     3.3
0    7.0
dtype: float64

[24]:

# without headers
import numpy
performance_weight = numpy.array(performance) * numpy.array(weight)
print(performance_weight)
weighted_average = numpy.sum(performance_weight)
print(weighted_average)

[[0.9 2.8 3.3]]
7.0

Functions

[25]:

import numpy
# the weighted average but with a function
def weighted_average_function(performance, weight):
    return numpy.sum(numpy.array(performance) * numpy.array(weight))
performance = [3,7,11]
weight = [.3,.4,.3]
weighted_average = weighted_average_function(performance, weight)
weighted_average

[25]:

7.0

Historical data

This page contains some samples for downloading historical data of your quant trading system.

Yahoo Finance API

There are many libraries for downloading historical data of stock markets from Yahoo finance, you can try some of them to discovery which you prefer:

Exchange rates (forex) from Yahoo finance
[1]:

!pip install pandas_datareader
from pandas_datareader import data as pdr
cross_from = ['EUR', 'USD', 'CAD']
cross_to = ['USD', 'JPY', 'USD']
symbols = [x + y + '=X' for x, y in zip(cross_from, cross_to)]
data = pdr.DataReader(symbols, 'yahoo')
data

Requirement already satisfied: pandas_datareader in /opt/conda/lib/python3.8/site-packages (0.9.0)
Requirement already satisfied: requests>=2.19.0 in /opt/conda/lib/python3.8/site-packages (from pandas_datareader) (2.25.1)
Requirement already satisfied: lxml in /opt/conda/lib/python3.8/site-packages (from pandas_datareader) (4.6.2)
Requirement already satisfied: pandas>=0.23 in /opt/conda/lib/python3.8/site-packages (from pandas_datareader) (1.1.5)
Requirement already satisfied: python-dateutil>=2.7.3 in /opt/conda/lib/python3.8/site-packages (from pandas>=0.23->pandas_datareader) (2.8.1)
Requirement already satisfied: pytz>=2017.2 in /opt/conda/lib/python3.8/site-packages (from pandas>=0.23->pandas_datareader) (2020.5)
Requirement already satisfied: numpy>=1.15.4 in /opt/conda/lib/python3.8/site-packages (from pandas>=0.23->pandas_datareader) (1.19.4)
Requirement already satisfied: six>=1.5 in /opt/conda/lib/python3.8/site-packages (from python-dateutil>=2.7.3->pandas>=0.23->pandas_datareader) (1.15.0)
Requirement already satisfied: chardet<5,>=3.0.2 in /opt/conda/lib/python3.8/site-packages (from requests>=2.19.0->pandas_datareader) (4.0.0)
Requirement already satisfied: certifi>=2017.4.17 in /opt/conda/lib/python3.8/site-packages (from requests>=2.19.0->pandas_datareader) (2020.12.5)
Requirement already satisfied: idna<3,>=2.5 in /opt/conda/lib/python3.8/site-packages (from requests>=2.19.0->pandas_datareader) (2.10)
Requirement already satisfied: urllib3<1.27,>=1.21.1 in /opt/conda/lib/python3.8/site-packages (from requests>=2.19.0->pandas_datareader) (1.26.2)

[1]:
Attributes Adj Close Close High Low Open Volume
Symbols EURUSD=X USDJPY=X CADUSD=X EURUSD=X USDJPY=X CADUSD=X EURUSD=X USDJPY=X CADUSD=X EURUSD=X USDJPY=X CADUSD=X EURUSD=X USDJPY=X CADUSD=X EURUSD=X USDJPY=X CADUSD=X
Date
2016-01-13 1.084399 117.830002 0.701508 1.084399 117.830002 0.701508 1.088400 118.362999 0.704871 1.080600 117.779999 0.700334 1.084399 117.831001 0.701459 0.0 0.0 0.0
2016-01-14 1.088601 117.425003 0.696816 1.088601 117.425003 0.696816 1.095000 118.161003 0.697515 1.083999 117.309998 0.694734 1.088803 117.400002 0.696811 0.0 0.0 0.0
2016-01-15 1.085894 118.203003 0.697126 1.085894 118.203003 0.697126 1.098200 118.255997 0.697253 1.085776 116.650002 0.687474 1.085776 118.223000 0.697204 0.0 0.0 0.0
2016-01-18 1.092001 116.989998 0.685965 1.092001 116.989998 0.685965 1.092705 117.433998 0.690274 1.087600 116.931000 0.685857 1.091906 117.010002 0.685857 0.0 0.0 0.0
2016-01-19 1.089503 117.384003 0.687545 1.089503 117.384003 0.687545 1.091000 118.099998 0.692953 1.086200 117.266998 0.687238 1.089503 117.397003 0.687564 0.0 0.0 0.0
... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
2021-01-05 1.225160 103.125000 0.782411 1.225160 103.125000 0.782411 1.229483 103.180000 0.787402 1.224995 102.671997 0.781959 1.225295 103.141998 0.782473 0.0 0.0 0.0
2021-01-06 1.230027 102.678001 0.788973 1.230027 102.678001 0.788973 1.235025 103.431000 0.791766 1.226693 102.589996 0.786250 1.229861 102.699997 0.788868 0.0 0.0 0.0
2021-01-07 1.234111 103.024002 0.789553 1.234111 103.024002 0.789553 1.234568 103.950996 0.789640 1.224665 102.956001 0.785379 1.233776 103.028000 0.789472 0.0 0.0 0.0
2021-01-08 1.227144 103.790001 0.788407 1.227144 103.790001 0.788407 1.228215 104.078003 0.789952 1.221493 103.616997 0.786201 1.226873 103.794998 0.788308 0.0 0.0 0.0
2021-01-11 1.216841 104.209999 0.782411 1.216841 104.209999 0.782411 1.223990 104.249001 0.788644 1.215806 103.686996 0.782142 1.222643 103.948997 0.788277 0.0 0.0 0.0

1281 rows × 18 columns

Other asset classes from Yahoo finance
[2]:

symbol = 'AMZN'
start = '1997-05-15'
end = '2020-09-30'
period = 'daily'

[3]:

from pandas_datareader import data as pdr
data = pdr.DataReader(symbol, 'yahoo', start, end)
len(data) # 5884 # until 2020-09-30 included
#data.to_csv('pdr_data.csv')
data

[3]:
High Low Open Close Volume Adj Close
Date
1997-05-15 2.500000 1.927083 2.437500 1.958333 72156000.0 1.958333
1997-05-16 1.979167 1.708333 1.968750 1.729167 14700000.0 1.729167
1997-05-19 1.770833 1.625000 1.760417 1.708333 6106800.0 1.708333
1997-05-20 1.750000 1.635417 1.729167 1.635417 5467200.0 1.635417
1997-05-21 1.645833 1.375000 1.635417 1.427083 18853200.0 1.427083
... ... ... ... ... ... ...
2020-09-24 3069.300049 2965.000000 2977.790039 3019.790039 5529400.0 3019.790039
2020-09-25 3101.540039 2999.000000 3054.860107 3095.129883 4615200.0 3095.129883
2020-09-28 3175.040039 3117.169922 3148.850098 3174.050049 4224200.0 3174.050049
2020-09-29 3188.260010 3132.540039 3175.389893 3144.879883 3495800.0 3144.879883
2020-09-30 3212.879883 3133.989990 3141.139893 3148.729980 4896100.0 3148.729980

5884 rows × 6 columns

[4]:

import pandas
!pip install yfinance
import yfinance
data = yfinance.Ticker(symbol)
data.info
df = pandas.DataFrame(data.history(period='max'))
len(df) # 5885 # until 2020-09-30 included + headline
#df.to_csv('yfinance_data.ticker.csv')
df

Requirement already satisfied: yfinance in /opt/conda/lib/python3.8/site-packages (0.1.55)
Requirement already satisfied: multitasking>=0.0.7 in /opt/conda/lib/python3.8/site-packages (from yfinance) (0.0.9)
Requirement already satisfied: lxml>=4.5.1 in /opt/conda/lib/python3.8/site-packages (from yfinance) (4.6.2)
Requirement already satisfied: numpy>=1.15 in /opt/conda/lib/python3.8/site-packages (from yfinance) (1.19.4)
Requirement already satisfied: requests>=2.20 in /opt/conda/lib/python3.8/site-packages (from yfinance) (2.25.1)
Requirement already satisfied: pandas>=0.24 in /opt/conda/lib/python3.8/site-packages (from yfinance) (1.1.5)
Requirement already satisfied: python-dateutil>=2.7.3 in /opt/conda/lib/python3.8/site-packages (from pandas>=0.24->yfinance) (2.8.1)
Requirement already satisfied: pytz>=2017.2 in /opt/conda/lib/python3.8/site-packages (from pandas>=0.24->yfinance) (2020.5)
Requirement already satisfied: six>=1.5 in /opt/conda/lib/python3.8/site-packages (from python-dateutil>=2.7.3->pandas>=0.24->yfinance) (1.15.0)
Requirement already satisfied: urllib3<1.27,>=1.21.1 in /opt/conda/lib/python3.8/site-packages (from requests>=2.20->yfinance) (1.26.2)
Requirement already satisfied: certifi>=2017.4.17 in /opt/conda/lib/python3.8/site-packages (from requests>=2.20->yfinance) (2020.12.5)
Requirement already satisfied: chardet<5,>=3.0.2 in /opt/conda/lib/python3.8/site-packages (from requests>=2.20->yfinance) (4.0.0)
Requirement already satisfied: idna<3,>=2.5 in /opt/conda/lib/python3.8/site-packages (from requests>=2.20->yfinance) (2.10)

[4]:
Open High Low Close Volume Dividends Stock Splits
Date
1997-05-15 2.437500 2.500000 1.927083 1.958333 72156000 0 0.0
1997-05-16 1.968750 1.979167 1.708333 1.729167 14700000 0 0.0
1997-05-19 1.760417 1.770833 1.625000 1.708333 6106800 0 0.0
1997-05-20 1.729167 1.750000 1.635417 1.635417 5467200 0 0.0
1997-05-21 1.635417 1.645833 1.375000 1.427083 18853200 0 0.0
... ... ... ... ... ... ... ...
2021-01-04 3270.000000 3272.000000 3144.020020 3186.629883 4411400 0 0.0
2021-01-05 3166.010010 3223.379883 3165.060059 3218.510010 2655500 0 0.0
2021-01-06 3146.479980 3197.510010 3131.159912 3138.379883 4394800 0 0.0
2021-01-07 3157.000000 3208.540039 3155.000000 3162.159912 3514500 0 0.0
2021-01-08 3180.000000 3190.639893 3142.199951 3182.699951 3534300 0 0.0

5953 rows × 7 columns

[5]:

data = yfinance.download(symbol, start=start, end=end, group_by='ticker')
len(data) # 5883 # until 2020-09-30 excluded
#data.to_csv('yfinance_data.download.csv')
data

[*********************100%***********************]  1 of 1 completed

[5]:
Open High Low Close Adj Close Volume
Date
1997-05-15 2.437500 2.500000 1.927083 1.958333 1.958333 72156000
1997-05-16 1.968750 1.979167 1.708333 1.729167 1.729167 14700000
1997-05-19 1.760417 1.770833 1.625000 1.708333 1.708333 6106800
1997-05-20 1.729167 1.750000 1.635417 1.635417 1.635417 5467200
1997-05-21 1.635417 1.645833 1.375000 1.427083 1.427083 18853200
... ... ... ... ... ... ...
2020-09-23 3120.429932 3127.000000 2992.379883 2999.860107 2999.860107 5652700
2020-09-24 2977.790039 3069.300049 2965.000000 3019.790039 3019.790039 5529400
2020-09-25 3054.860107 3101.540039 2999.000000 3095.129883 3095.129883 4615200
2020-09-28 3148.850098 3175.040039 3117.169922 3174.050049 3174.050049 4224200
2020-09-29 3175.389893 3188.260010 3132.540039 3144.879883 3144.879883 3495800

5883 rows × 6 columns

[6]:

yfinance.pdr_override()
data = pdr.get_data_yahoo(symbol, start=start, end=end)
len(data) # 5883 # until 2020-09-30 excluded
#data.to_csv('yfpdr_data.csv')
data

[*********************100%***********************]  1 of 1 completed

[6]:
Open High Low Close Adj Close Volume
Date
1997-05-15 2.437500 2.500000 1.927083 1.958333 1.958333 72156000
1997-05-16 1.968750 1.979167 1.708333 1.729167 1.729167 14700000
1997-05-19 1.760417 1.770833 1.625000 1.708333 1.708333 6106800
1997-05-20 1.729167 1.750000 1.635417 1.635417 1.635417 5467200
1997-05-21 1.635417 1.645833 1.375000 1.427083 1.427083 18853200
... ... ... ... ... ... ...
2020-09-23 3120.429932 3127.000000 2992.379883 2999.860107 2999.860107 5652700
2020-09-24 2977.790039 3069.300049 2965.000000 3019.790039 3019.790039 5529400
2020-09-25 3054.860107 3101.540039 2999.000000 3095.129883 3095.129883 4615200
2020-09-28 3148.850098 3175.040039 3117.169922 3174.050049 3174.050049 4224200
2020-09-29 3175.389893 3188.260010 3132.540039 3144.879883 3144.879883 3495800

5883 rows × 6 columns

[7]:

!pip install yahoofinancials
from yahoofinancials import YahooFinancials
data = YahooFinancials(symbol)
json = data.get_historical_price_data(start_date=start, end_date=end, time_interval=period)
df = pandas.DataFrame(json[symbol]['prices'])
len(df) # 5883 # until 2020-09-30 excluded and seconds instead of date
#df.to_csv('yahoofinancials_data.csv')
df

Requirement already satisfied: yahoofinancials in /opt/conda/lib/python3.8/site-packages (1.6)
Requirement already satisfied: beautifulsoup4 in /opt/conda/lib/python3.8/site-packages (from yahoofinancials) (4.9.3)
Requirement already satisfied: pytz in /opt/conda/lib/python3.8/site-packages (from yahoofinancials) (2020.5)
Requirement already satisfied: soupsieve>1.2 in /opt/conda/lib/python3.8/site-packages (from beautifulsoup4->yahoofinancials) (2.0.1)

[7]:
date high low open close volume adjclose formatted_date
0 863703000 2.500000 1.927083 2.437500 1.958333 72156000 1.958333 1997-05-15
1 863789400 1.979167 1.708333 1.968750 1.729167 14700000 1.729167 1997-05-16
2 864048600 1.770833 1.625000 1.760417 1.708333 6106800 1.708333 1997-05-19
3 864135000 1.750000 1.635417 1.729167 1.635417 5467200 1.635417 1997-05-20
4 864221400 1.645833 1.375000 1.635417 1.427083 18853200 1.427083 1997-05-21
... ... ... ... ... ... ... ... ...
5878 1600867800 3127.000000 2992.379883 3120.429932 2999.860107 5652700 2999.860107 2020-09-23
5879 1600954200 3069.300049 2965.000000 2977.790039 3019.790039 5529400 3019.790039 2020-09-24
5880 1601040600 3101.540039 2999.000000 3054.860107 3095.129883 4615200 3095.129883 2020-09-25
5881 1601299800 3175.040039 3117.169922 3148.850098 3174.050049 4224200 3174.050049 2020-09-28
5882 1601386200 3188.260010 3132.540039 3175.389893 3144.879883 3495800 3144.879883 2020-09-29

5883 rows × 8 columns

[8]:

!pip install yahoo_fin
from yahoo_fin import stock_info as si
data = si.get_data(symbol)
len(data) # 5885 # until 2020-09-30 included (+ headline)
#data.to_csv('yahoo_fin_data.csv')
data

Requirement already satisfied: yahoo_fin in /opt/conda/lib/python3.8/site-packages (0.8.6)
Warning - Certain functionality
             requires requests_html, which is not installed.

             Install using:
             pip install requests_html

             After installation, you may have to restart your Python session.

[8]:
open high low close adjclose volume ticker
1997-05-15 2.437500 2.500000 1.927083 1.958333 1.958333 72156000 AMZN
1997-05-16 1.968750 1.979167 1.708333 1.729167 1.729167 14700000 AMZN
1997-05-19 1.760417 1.770833 1.625000 1.708333 1.708333 6106800 AMZN
1997-05-20 1.729167 1.750000 1.635417 1.635417 1.635417 5467200 AMZN
1997-05-21 1.635417 1.645833 1.375000 1.427083 1.427083 18853200 AMZN
... ... ... ... ... ... ... ...
2021-01-04 3270.000000 3272.000000 3144.020020 3186.629883 3186.629883 4411400 AMZN
2021-01-05 3166.010010 3223.379883 3165.060059 3218.510010 3218.510010 2655500 AMZN
2021-01-06 3146.479980 3197.510010 3131.159912 3138.379883 3138.379883 4394800 AMZN
2021-01-07 3157.000000 3208.540039 3155.000000 3162.159912 3162.159912 3514500 AMZN
2021-01-08 3180.000000 3190.639893 3142.199951 3182.699951 3182.699951 3534300 AMZN

5953 rows × 7 columns

Exchange rates (forex) from Oanda

Oanda API allows to download the historical data with the API key of a free Oanda account: you can find the Oanda library or you can use many other backtesting libraries for this.

But this is a general tutorial, so you can find the data by oanda.URL that the getSymbols method of R quantmod library uses. When you use this method, you can only download the historical data of the last six months.

[9]:

import json
import pandas
import requests
from datetime import datetime, timedelta
base_url = 'https://www.oanda.com/fx-for-business/historical-rates/api/data/update/?&source=OANDA&adjustment=0&base_currency={to_currency}&start_date={start_date}&end_date={end_date}&period={period}&price=mid&view=table&quote_currency_0={from_currency}'
today = datetime.strftime(datetime.now(), '%Y-%m-%d')
def get_oanda_currency_historical_rates(base_url, start, end, quote_currency, base_currency, period):
    url = base_url.format(from_currency=quote_currency, to_currency=base_currency, start_date=start, end_date=end, period=period)
    response = json.loads(requests.get(url).content.decode('utf-8'))
    return pandas.DataFrame(response['widget'][0]['data'])
# the last 6 months
df = get_oanda_currency_historical_rates(base_url, datetime.strftime(datetime.now() - timedelta(180), '%Y-%m-%d'), today, 'EUR', 'USD', period)
df

[9]:
0 1
0 1610236800000 0.818336
1 1610150400000 0.818270
2 1610064000000 0.816854
3 1609977600000 0.813786
4 1609891200000 0.812080
... ... ...
175 1595116800000 0.875000
176 1595030400000 0.875028
177 1594944000000 0.876130
178 1594857600000 0.877077
179 1594771200000 0.875957

180 rows × 2 columns

[10]:

# the last 60 days
df = get_oanda_currency_historical_rates(base_url, datetime.strftime(datetime.now() - timedelta(60), '%Y-%m-%d'), today, 'EUR', 'USD', period)
df.head()

[10]:
0 1
0 1610236800000 0.818336
1 1610150400000 0.818270
2 1610064000000 0.816854
3 1609977600000 0.813786
4 1609891200000 0.812080

Crypto currencies from CoinMarketCap

CoinMarketCap API allows to download the historical data with the API key of a free CoinMarketCap account: you can find the CoinMarketCap library or you can use many other backtesting libraries for this.

Like for the all assets, there are many libraries that they implement Coinmarketcap API, but many are open source so the updates could be slowly:

Below you can find a sample with sandbox API of CoinMarketCap.

[11]:

import json
import pandas
import requests
base_url = 'https://sandbox-api.coinmarketcap.com/v1/'
def get_coinmarketcap_data(url):
    response = json.loads(requests.get(url).content.decode('utf-8'))
    return pandas.DataFrame(response['data'])

# returns all active cryptocurrencies
url = base_url + 'cryptocurrency/map'
marketcap_map = get_coinmarketcap_data(url)
marketcap_map

[11]:
id name symbol slug is_active rank platform
0 1 Bitcoin BTC bitcoin 1 1 None
1 2 Litecoin LTC litecoin 1 5 None
2 3 Namecoin NMC namecoin 1 310 None
3 4 Terracoin TRC terracoin 1 926 None
4 5 Peercoin PPC peercoin 1 346 None
... ... ... ... ... ... ... ...
2322 4266 Monarch MT monarch 1 2059 {'id': 1027, 'name': 'Ethereum', 'symbol': 'ET...
2323 4268 NewYork Exchange NYE newyork-exchange 1 2012 None
2324 4273 Sessia KICKS sessia 1 2118 {'id': 1027, 'name': 'Ethereum', 'symbol': 'ET...
2325 4275 Cocos-BCX COCOS cocos-bcx 1 1960 {'id': 1027, 'name': 'Ethereum', 'symbol': 'ET...
2326 4276 Defi DEFI defi 1 1965 {'id': 1027, 'name': 'Ethereum', 'symbol': 'ET...

2327 rows × 7 columns

[12]:

# returns last market data of all active cryptocurrencies
url = base_url + 'cryptocurrency/listings/latest?convert=EUR'
last_marketcap = get_coinmarketcap_data(url)
last_marketcap

[12]:
id name symbol slug num_market_pairs date_added tags max_supply circulating_supply total_supply platform cmc_rank last_updated quote
0 1 Bitcoin BTC bitcoin 7919 2013-04-28T00:00:00.000Z [mineable] 2.100000e+07 1.790601e+07 1.790601e+07 None 1 2019-08-30T18:51:28.000Z {'EUR': {'price': 8708.442730269642, 'volume_2...
1 1027 Ethereum ETH ethereum 5629 2015-08-07T00:00:00.000Z [mineable] NaN 1.075379e+08 1.075379e+08 None 2 2019-08-30T18:51:21.000Z {'EUR': {'price': 153.68597253794135, 'volume_...
2 52 XRP XRP ripple 449 2013-08-04T00:00:00.000Z [] 1.000000e+11 4.293287e+10 9.999137e+10 None 3 2019-08-30T18:51:03.000Z {'EUR': {'price': 0.23181863120417767, 'volume...
3 1831 Bitcoin Cash BCH bitcoin-cash 378 2017-07-23T00:00:00.000Z [mineable] 2.100000e+07 1.797598e+07 1.797598e+07 None 4 2019-08-30T18:51:08.000Z {'EUR': {'price': 256.03512635975414, 'volume_...
4 2 Litecoin LTC litecoin 538 2013-04-28T00:00:00.000Z [mineable] 8.400000e+07 6.314712e+07 6.314712e+07 None 5 2019-08-30T18:51:04.000Z {'EUR': {'price': 58.62607480610576, 'volume_2...
... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
95 3709 Grin GRIN grin 53 2019-01-27T00:00:00.000Z [mineable] NaN 1.954596e+07 1.954596e+07 None 96 2019-08-30T18:51:18.000Z {'EUR': {'price': 1.7584409339670577, 'volume_...
96 2694 Nexo NEXO nexo 22 2018-05-01T00:00:00.000Z [] NaN 5.600000e+08 1.000000e+09 {'id': 1027, 'name': 'Ethereum', 'symbol': 'ET... 97 2019-08-30T18:51:12.000Z {'EUR': {'price': 0.06125607986488225, 'volume...
97 2900 Project Pai PAI project-pai 17 2018-07-05T00:00:00.000Z [] NaN 1.451234e+09 1.618050e+09 None 98 2019-08-30T18:51:14.000Z {'EUR': {'price': 0.022456686270928363, 'volum...
98 2137 Electroneum ETN electroneum 20 2017-11-02T00:00:00.000Z [mineable] 2.100000e+10 9.792946e+09 9.792946e+09 None 99 2019-08-30T18:51:08.000Z {'EUR': {'price': 0.0032193992228962184, 'volu...
99 1681 PRIZM PZM prizm 7 2017-05-19T00:00:00.000Z [] 6.000000e+09 9.820737e+07 9.820737e+07 None 100 2019-08-30T18:51:05.000Z {'EUR': {'price': 0.3204834704804656, 'volume_...

100 rows × 14 columns

[13]:

# returns last market data of Bitcoin
url = base_url + 'cryptocurrency/quotes/latest?id=1&convert=EUR'
last_bitcoin = get_coinmarketcap_data(url)
last_bitcoin

[13]:
1
circulating_supply 17906012
cmc_rank 1
date_added 2013-04-28T00:00:00.000Z
id 1
is_active 1
is_fiat 0
is_market_cap_included_in_calc 1
last_updated 2019-08-30T18:51:28.000Z
max_supply 21000000
name Bitcoin
num_market_pairs 7919
platform None
quote {'EUR': {'price': 8708.442730269642, 'volume_2...
slug bitcoin
symbol BTC
tags [mineable]
total_supply 17906012 
[14]:

# returns historical data from a time_start to a time_end
url = base_url + 'cryptocurrency/quotes/historical?id=1&convert=EUR&time_start=2020-05-21T12:14&time_end=2020-05-21T12:44'
data = get_coinmarketcap_data(url)
data

[14]:
id name symbol is_fiat quotes
0 1 Bitcoin BTC 0 {'timestamp': '2020-05-21T12:19:03.000Z', 'quo...
1 1 Bitcoin BTC 0 {'timestamp': '2020-05-21T12:24:02.000Z', 'quo...
2 1 Bitcoin BTC 0 {'timestamp': '2020-05-21T12:29:03.000Z', 'quo...
3 1 Bitcoin BTC 0 {'timestamp': '2020-05-21T12:34:03.000Z', 'quo...
4 1 Bitcoin BTC 0 {'timestamp': '2020-05-21T12:39:01.000Z', 'quo...
5 1 Bitcoin BTC 0 {'timestamp': '2020-05-21T12:44:02.000Z', 'quo...

Other asset classes from Quandl

You can see details of asset classes of Quandl, by its research, for example crude oil.

Quandl API allows to download the historical data with the API key of a free Quandl account: you can find the Quandl library (doc) or you can use many other backtesting libraries for this.

However, it is possible to execute 50 calls per day without API key.

[15]:

symbol = 'CHRIS/CME_QM1'

import pandas
from datetime import datetime
base_url = 'https://www.quandl.com/api/v3/datasets/{symbol}.csv?start_date={start_date}&end_date={end_date}&collapse={period}'
today = datetime.strftime(datetime.now(), '%Y-%m-%d')
def get_quandl_financial_data(base_url, symbol, start, end, period):
    url = base_url.format(symbol=symbol, start_date=start, end_date=end, period=period)
    return pandas.read_csv(url)
# returns all free columns values
df = get_quandl_financial_data(base_url, symbol, start, end, period)
df

[15]:
Date Open High Low Last Change Settle Volume Previous Day Open Interest
0 2020-09-30 39.225 40.375 38.675 39.850 0.93 40.22 10605.0 1252.0
1 2020-09-29 40.575 40.700 38.425 39.075 -1.31 39.29 12589.0 784.0
2 2020-09-28 40.125 40.800 39.775 40.600 0.35 40.60 8194.0 787.0
3 2020-09-25 40.175 40.625 39.700 40.075 -0.06 40.25 7024.0 879.0
4 2020-09-24 39.600 40.350 39.125 40.200 0.38 40.31 8349.0 869.0
... ... ... ... ... ... ... ... ... ...
1646 2014-03-12 99.500 99.575 97.550 97.990 2.04 97.99 9437.0 2462.0
1647 2014-03-11 100.925 101.500 99.350 100.030 1.09 100.03 7337.0 2300.0
1648 2014-03-07 101.900 102.900 101.575 102.580 1.02 102.58 5304.0 2315.0
1649 2014-03-06 101.000 102.050 100.150 101.560 0.11 101.56 6850.0 2383.0
1650 2014-03-05 103.325 103.525 100.825 101.450 1.88 101.45 7459.0 2305.0

1651 rows × 9 columns

[16]:

# returns only open values
df['Open'] # or
df = get_quandl_financial_data(base_url + '&column_index=1', symbol, start, end, period)
df

[16]:
Date Open
0 2020-09-30 39.225
1 2020-09-29 40.575
2 2020-09-28 40.125
3 2020-09-25 40.175
4 2020-09-24 39.600
... ... ...
1646 2014-03-12 99.500
1647 2014-03-11 100.925
1648 2014-03-07 101.900
1649 2014-03-06 101.000
1650 2014-03-05 103.325

1651 rows × 2 columns

[17]:

!pip install quandl
import quandl
# returns all free columns values
data = quandl.get(symbol)
data

Requirement already satisfied: quandl in /opt/conda/lib/python3.8/site-packages (3.5.3)
Requirement already satisfied: requests>=2.7.0 in /opt/conda/lib/python3.8/site-packages (from quandl) (2.25.1)
Requirement already satisfied: six in /opt/conda/lib/python3.8/site-packages (from quandl) (1.15.0)
Requirement already satisfied: python-dateutil in /opt/conda/lib/python3.8/site-packages (from quandl) (2.8.1)
Requirement already satisfied: inflection>=0.3.1 in /opt/conda/lib/python3.8/site-packages (from quandl) (0.5.1)
Requirement already satisfied: more-itertools in /opt/conda/lib/python3.8/site-packages (from quandl) (8.6.0)
Requirement already satisfied: numpy>=1.8 in /opt/conda/lib/python3.8/site-packages (from quandl) (1.19.4)
Requirement already satisfied: pandas>=0.14 in /opt/conda/lib/python3.8/site-packages (from quandl) (1.1.5)
Requirement already satisfied: pytz>=2017.2 in /opt/conda/lib/python3.8/site-packages (from pandas>=0.14->quandl) (2020.5)
Requirement already satisfied: certifi>=2017.4.17 in /opt/conda/lib/python3.8/site-packages (from requests>=2.7.0->quandl) (2020.12.5)
Requirement already satisfied: urllib3<1.27,>=1.21.1 in /opt/conda/lib/python3.8/site-packages (from requests>=2.7.0->quandl) (1.26.2)
Requirement already satisfied: chardet<5,>=3.0.2 in /opt/conda/lib/python3.8/site-packages (from requests>=2.7.0->quandl) (4.0.0)
Requirement already satisfied: idna<3,>=2.5 in /opt/conda/lib/python3.8/site-packages (from requests>=2.7.0->quandl) (2.10)

[17]:
Open High Low Last Change Settle Volume Previous Day Open Interest
Date
2014-03-05 103.325 103.525 100.825 101.450 1.88 101.45 7459.0 2305.0
2014-03-06 101.000 102.050 100.150 101.560 0.11 101.56 6850.0 2383.0
2014-03-07 101.900 102.900 101.575 102.580 1.02 102.58 5304.0 2315.0
2014-03-11 100.925 101.500 99.350 100.030 1.09 100.03 7337.0 2300.0
2014-03-12 99.500 99.575 97.550 97.990 2.04 97.99 9437.0 2462.0
... ... ... ... ... ... ... ... ...
2021-01-04 48.400 49.850 47.200 47.350 -0.90 47.62 21845.0 1350.0
2021-01-05 47.400 50.200 47.275 49.850 2.31 49.93 23473.0 1799.0
2021-01-06 49.825 50.925 49.475 50.525 0.70 50.63 16675.0 1908.0
2021-01-07 50.525 51.275 50.400 50.950 0.20 50.83 10814.0 1800.0
2021-01-08 50.925 52.750 50.825 52.750 1.41 52.24 10770.0 1833.0

1720 rows × 8 columns

[18]:

# returns only open values
data['Open']

[18]:

Date
2014-03-05    103.325
2014-03-06    101.000
2014-03-07    101.900
2014-03-11    100.925
2014-03-12     99.500
               ...
2021-01-04     48.400
2021-01-05     47.400
2021-01-06     49.825
2021-01-07     50.525
2021-01-08     50.925
Name: Open, Length: 1720, dtype: float64

Graphics

This is a sample for plotting with Python. There are 2 principal libraries for plotting,

  • the library matplotlib that it has many utilities like mplfinance for the visualization, and visual analysis, of financial data

  • the library plotly that it is a very interesting tool makes interactive and publication-quality graphs

With pandas_datareader library you can download the historical data and with pandas library you can make the data range.

[1]:

# initialization
import pandas as pd
from pandas_datareader import data as pdr
start='2017-10-30'
end='2020-10-08'
amzn = pdr.DataReader('AMZN', 'yahoo', start, end)
# simple indicators
amzn['SMA25'] = amzn['Close'].rolling(25).mean()
amzn['EMA25'] = amzn['Close'].ewm(span=25, adjust=False).mean()
amzn['STD25'] = amzn['Close'].rolling(25).std()
amzn['Upper Band'] = amzn['SMA25'] + (amzn['STD25'] * 2)
amzn['Lower Band'] = amzn['SMA25'] - (amzn['STD25'] * 2)
# data ranges
all_bm = pd.date_range(start=start, end=end, freq='BM')
all_range = pd.date_range(start=start, end=end)
amzn_bm = amzn.reindex(all_bm)
amzn_bm_all = amzn_bm.reindex(all_range)
amzn_all = amzn.reindex(all_range)
piece_d = pd.date_range(start='2020-01-01', end='2020-10-01')
amzn_bm_piece_d = amzn_bm.reindex(piece_d)
amzn_piece_d = amzn.reindex(piece_d)
piece_bm = pd.date_range(start='2020-01-01', end='2020-10-01', freq='BM')
amzn_bm_piece_bm = amzn_bm.reindex(piece_bm)

Matplotlib is very powerful: you can add tracks with different date range.

[2]:

# plot with daily, monthly and sma 25
import matplotlib.pyplot as plt
fig, ax = plt.subplots(figsize=(16,9))
ax.plot(amzn['Close'].index, amzn['Close'], label='AMZN daily')
ax.plot(amzn['SMA25'].index, amzn['SMA25'], label='AMZN sma25')
ax.plot(amzn_bm['Close'].index, amzn_bm['Close'], label='AMZN monthly')
ax.set_xlabel('Date')
ax.set_ylabel('AMZN ($)')
ax.legend()
plt.show()
#plt.savefig('plot.with.daily.monthly.sma.25.png')
_images/python_graphics_3_0.png

mplfinance contains the matplotlib finance API that makes it easier to create financial plots.

[3]:

# plot with candle daily and sma 25
!pip install mplfinance
import mplfinance as mpf
kwargs = dict(type='candle',mav=(25),volume=True,figratio=(16,9),figscale=1)
mpf.plot(amzn,**kwargs)
#plt.savefig('plot.with.candle.daily.sma.25.png')

Requirement already satisfied: mplfinance in /opt/conda/lib/python3.8/site-packages (0.12.7a4)
Requirement already satisfied: pandas in /opt/conda/lib/python3.8/site-packages (from mplfinance) (1.1.5)
Requirement already satisfied: matplotlib in /opt/conda/lib/python3.8/site-packages (from mplfinance) (3.3.3)
Requirement already satisfied: kiwisolver>=1.0.1 in /opt/conda/lib/python3.8/site-packages (from matplotlib->mplfinance) (1.3.1)
Requirement already satisfied: cycler>=0.10 in /opt/conda/lib/python3.8/site-packages (from matplotlib->mplfinance) (0.10.0)
Requirement already satisfied: pillow>=6.2.0 in /opt/conda/lib/python3.8/site-packages (from matplotlib->mplfinance) (8.0.1)
Requirement already satisfied: numpy>=1.15 in /opt/conda/lib/python3.8/site-packages (from matplotlib->mplfinance) (1.19.4)
Requirement already satisfied: python-dateutil>=2.1 in /opt/conda/lib/python3.8/site-packages (from matplotlib->mplfinance) (2.8.1)
Requirement already satisfied: pyparsing!=2.0.4,!=2.1.2,!=2.1.6,>=2.0.3 in /opt/conda/lib/python3.8/site-packages (from matplotlib->mplfinance) (2.4.7)
Requirement already satisfied: six in /opt/conda/lib/python3.8/site-packages (from cycler>=0.10->matplotlib->mplfinance) (1.15.0)
Requirement already satisfied: pytz>=2017.2 in /opt/conda/lib/python3.8/site-packages (from pandas->mplfinance) (2020.5)
_images/python_graphics_5_1.png

There are many styles to customize your graphs.

[4]:

# plot with candle daily, sma 25 and yahoo style
mpf.plot(amzn,**kwargs,style='yahoo')
#plt.savefig('plot.with.candle.daily.sma.25.yahoo.style.png')
_images/python_graphics_7_0.png

You can add an array of plots that each has to be a dateframe.

[5]:

# plot with candle daily, monthly points, sma 25 and yahoo style
aps = [
    mpf.make_addplot(amzn_all['SMA25']),
    mpf.make_addplot(amzn_bm_all['Close'],type='scatter')
]
mpf.plot(amzn_all,**kwargs,style='yahoo',addplot=aps)
#plt.savefig('plot.with.candle.daily.monthly.points.sma.25.yahoo.style.png')
_images/python_graphics_9_0.png

You can enlarge a window to observe that the weeks are not contiguous.

[6]:

# plot with candle daily, monthly points, sma 25 and yahoo style on a little date range
aps = [
    mpf.make_addplot(amzn_piece_d['SMA25']),
    mpf.make_addplot(amzn_bm_piece_d['Close'],type='scatter')
]
mpf.plot(amzn_piece_d,**kwargs,style='yahoo',addplot=aps)
#plt.savefig('plot.with.candle.daily.monthly.points.sma.25.yahoo.style.little.range.png')
_images/python_graphics_11_0.png

You can add an array of simple lines date-value or other specific lines for trends, support, resistance, and trading.

[7]:

# plot with candle daily, monthly points and monthly line, sma 25 and yahoo style on a little date range
subset = pd.DataFrame(amzn_bm_piece_bm['Close'])
amzn_bm_line = list(subset.itertuples(index=True, name=None))
aps = [
    mpf.make_addplot(amzn_piece_d['SMA25']),
    mpf.make_addplot(amzn_bm_piece_d['Close'],type='scatter',markersize=78)
]
mpf.plot(amzn_piece_d,**kwargs,style='yahoo',addplot=aps,alines=amzn_bm_line)
#plt.savefig('plot.with.candle.daily.monthly.points.monthly.line.sma.25.yahoo.style.little.range.png')
_images/python_graphics_13_0.png

You can also plot with matplotlib style for having more customization like legend.

[8]:

# the previous plot with matplotlib style plus old finance modules
import matplotlib.dates as mdates
from matplotlib.dates import MONDAY, DateFormatter, DayLocator, WeekdayLocator
from mplfinance.original_flavor import candlestick_ohlc

# configurations
fig, ax = plt.subplots(figsize=(16,9))

mondays = WeekdayLocator(MONDAY)        # major ticks on the mondays
alldays = DayLocator()              # minor ticks on the days
weekFormatter = DateFormatter('%b %d')  # e.g., Jan 12
dayFormatter = DateFormatter('%d')      # e.g., 12
ax.xaxis.set_major_locator(mondays)
#ax.xaxis.set_minor_locator(alldays)
ax.xaxis.set_major_formatter(weekFormatter)
#ax.xaxis.set_minor_formatter(dayFormatter)
plt.setp(plt.gca().get_xticklabels(), rotation=45, horizontalalignment='right')

# grid
plt.rc('axes', grid=True)

# the plots
candlestick_ohlc(ax, zip(mdates.date2num(amzn_piece_d.index.to_pydatetime()),
                 amzn_piece_d['Open'], amzn_piece_d['High'],amzn_piece_d['Low'], amzn_piece_d['Close']),
                 colorup='g', colordown='r', width=0.6)
ax.plot(amzn_bm_piece_bm['Close'].index, amzn_bm_piece_bm['Close'], label='AMZN monthly')
ax.plot(amzn_bm_piece_bm['Close'].index, amzn_bm_piece_bm['Close'], label='AMZN marker', marker='o', markersize=16, color='C0')
ax.plot(amzn_piece_d['SMA25'].index, amzn_piece_d['SMA25'], label='AMZN sma25')

# other configurations
ax.set_title('AMZN')
#ax.set_xlabel('Date')
ax.set_ylabel('Price')
ax.legend()
plt.show()
#plt.savefig('plot.with.candle.daily.monthly.points.monthly.line.sma.25.matplotlib.style.little.range.png')
_images/python_graphics_15_0.png

Another customization could be to remove the gaps.

[9]:

# the previous plot with matplotlib style without gaps
import numpy as np

# configurations
fig, ax = plt.subplots(figsize=(16,9))

# remove the gaps
data = pd.DataFrame(amzn_piece_d.dropna())
# Preserve dates to be re-labelled later.
x_dates = data.index.to_pydatetime()
# Override data['date'] with a list of incrementatl integers.
#   This will not create gaps in the candle stick graph.
data_size = len(x_dates)
data['Date'] = np.arange(start = 0, stop = data_size, step = 1, dtype='int')
# Re-arrange so that each line contains values of a day: 'date','open','high','low','close'.
quotes = [tuple(x) for x in data[['Date','Open','High','Low','Close']].values]

# Go through each x-tick label and rename it.
x_tick_labels = []
x_tick_positions = []
x_bm_line = []
for l_date in x_dates:
    date_str = ''
#    if l_date.strftime('%A') == 'Monday':
    if l_date.strftime('%A') == 'Tuesday':
        date_str = l_date.strftime('%b %d')
        x_tick_labels.append(date_str)
        x_tick_positions.append(list(x_dates).index(l_date))
    for date in amzn_bm_piece_bm.index:
        if l_date == date:
            x_bm_line.append(list(x_dates).index(l_date))
ax.set(xticks=x_tick_positions, xticklabels=x_tick_labels)
plt.setp(plt.gca().get_xticklabels(), rotation=45, horizontalalignment='right')

# grid
plt.rc('axes', grid=True)

# the plots
candlestick_ohlc(ax, quotes, colorup='g', colordown='r', width=0.6)
ax.plot(x_bm_line, amzn_bm_piece_bm['Close'], label='AMZN monthly')
ax.plot(x_bm_line, amzn_bm_piece_bm['Close'], label='AMZN marker', marker='o', markersize=16, color='C0')
ax.plot(data['Date'], data['SMA25'], label='AMZN sma25')

# other configurations
ax.set_title('AMZN')
#ax.set_xlabel('Date')
ax.set_ylabel('Price')
ax.legend()
plt.show()
#plt.savefig('plot.with.candle.daily.monthly.points.monthly.line.sma.25.matplotlib.style.little.range.without.gaps.png')
_images/python_graphics_17_0.png

And to add the volume like bars.

[10]:

# the previous plot with matplotlib style without gaps plus volume

# configurations
fig, ax = plt.subplots(figsize=(16,9))
ax = plt.subplot2grid((3,3), (0,0), colspan=3, rowspan=2)
ax.yaxis.set_label_position('right')
ax.yaxis.tick_right()
ax2 = plt.subplot2grid((3,3), (2,0), colspan=3)
ax2.yaxis.set_label_position('right')
ax2.yaxis.tick_right()
#ax.axes.get_xaxis().set_visible(False)
fig.subplots_adjust(hspace=0)

# remove the gaps
data = pd.DataFrame(amzn_piece_d.dropna())
# Preserve dates to be re-labelled later.
x_dates = data.index.to_pydatetime()
# Override data['date'] with a list of incrementatl integers.
#   This will not create gaps in the candle stick graph.
data_size = len(x_dates)
data['Date'] = np.arange(start = 0, stop = data_size, step = 1, dtype='int')
# Re-arrange so that each line contains values of a day: 'date','open','high','low','close'.
quotes = [tuple(x) for x in data[['Date','Open','High','Low','Close']].values]

# Go through each x-tick label and rename it.
x_tick_labels_empty = []
x_tick_labels = []
x_tick_positions = []
x_bm_line = []
for l_date in x_dates:
    date_str = ''
#    if l_date.strftime('%A') == 'Monday':
    if l_date.strftime('%A') == 'Tuesday':
        date_str = l_date.strftime('%b %d')
        x_tick_labels.append(date_str)
        x_tick_labels_empty.append('')
        x_tick_positions.append(list(x_dates).index(l_date))
    for date in amzn_bm_piece_bm.index:
        if l_date == date:
            x_bm_line.append(list(x_dates).index(l_date))
#ax.set(xticks=x_tick_positions, xticklabels=x_tick_labels)
ax.set(xticks=x_tick_positions, xticklabels=x_tick_labels_empty)
ax2.set(xticks=x_tick_positions, xticklabels=x_tick_labels)
plt.setp(plt.gca().get_xticklabels(), rotation=45, horizontalalignment='right')

# grid
plt.rc('axes', grid=True)

# the plots
candlestick_ohlc(ax, quotes, colorup='g', colordown='r', width=0.6)
ax.plot(x_bm_line, amzn_bm_piece_bm['Close'], label='AMZN monthly')
ax.plot(x_bm_line, amzn_bm_piece_bm['Close'], label='AMZN marker', marker='o', markersize=16, color='C0')
ax.plot(data['Date'], data['SMA25'], label='AMZN sma25')

# make bar plots and color differently depending on up/down for the day
pos = data['Open']-data['Close']<0
neg = data['Open']-data['Close']>0
ax2.bar(data['Date'][pos],data['Volume'][pos],color='green',width=0.6,align='center')
ax2.bar(data['Date'][neg],data['Volume'][neg],color='red',width=0.6,align='center')

# other configurations
ax.set_title('AMZN')
#ax2.set_xlabel('Date')
ax.set_ylabel('Price')
ax2.set_ylabel('Volume x 1e7')
# avoid scientific notation (1e7)
ax2.get_yaxis().get_offset_text().set_visible(False)
ax.legend()
plt.show()
#plt.savefig('plot.with.candle.daily.monthly.points.monthly.line.sma.25.matplotlib.style.little.range.without.gaps.plus.volume.png')
_images/python_graphics_19_0.png

Matplotlib is an advanced library for plotting everything you want, but maybe at the beginning, you need only to add the standard indicators.

It is simple to add Simple Moving Averages (SMA) with mplfinance, but this library contains only this indicator.

There are many libraries that they are a copy of TA-lib, an historical library of Technical Analisys.

  • some libraries are a wrapper of the original TA-lib, like mrjbq7/ta-lib, and it needs the original TA-lib is installed

  • some libraries are a wrapper of the Matplotlib, like ricequant/rqalpha

  • many libraries create plots, calculate indicators and make backtesting starting from the previous ones: you only have to choose what you like best

[11]:

# plot with candle daily, sma 25, ema 25 and yahoo style without gaps
data = pd.DataFrame(amzn_piece_d.dropna())
aps = [
    mpf.make_addplot(data['SMA25'],color='C1'), # orange
    mpf.make_addplot(data['EMA25'],color='C3'), # red
    mpf.make_addplot(data['Upper Band'],linestyle='-.',color='g'),
    mpf.make_addplot(data['Lower Band'],linestyle='-.',color='g'),
]
mpf.plot(data,**kwargs,style='yahoo',addplot=aps,fill_between=dict(y1=data['Lower Band'].values,y2=data['Upper Band'].values,alpha=0.1,color='g'))
#plt.savefig('plot.with.candle.daily.sma.25.ema.25.yahoo.style.without.gaps.png')
_images/python_graphics_21_0.png

Strategies

You can initilize a strategy by Python language.

There are many libraries that you can use for defining your trading strategy: you only have to choose what you like best. In these samples, the library used for trading strategy is Backtesting.py.

How to load indicators on your strategy

The best practice is to prepare one column for each indicator that you want to use on your strategy.

[1]:

!pip install pandas_datareader
# initialization
import numpy as np
import pandas as pd
from pandas_datareader import data as pdr
start='2017-10-30'
end='2020-10-08'
amzn = pdr.DataReader('AMZN', 'yahoo', start, end)
# simple indicators
amzn['SMA20'] = amzn['Close'].rolling(20).mean()
amzn['SMA50'] = amzn['Close'].rolling(50).mean()
amzn['EMA20'] = amzn['Close'].ewm(span=20, adjust=False).mean()
amzn['EMA50'] = amzn['Close'].ewm(span=50, adjust=False).mean()
amzn['STD20'] = amzn['Close'].rolling(20).std()
amzn['Upper Band'] = amzn['SMA20'] + (amzn['STD20'] * 2)
amzn['Lower Band'] = amzn['SMA20'] - (amzn['STD20'] * 2)
#amzn['Upper Band'] = amzn['EMA20'] + (amzn['STD20'] * 2)
#amzn['Lower Band'] = amzn['EMA20'] - (amzn['STD20'] * 2)
def RSI(data, time_window):
    diff = data.diff(1).dropna()
    up_chg = 0 * diff
    down_chg = 0 * diff
    up_chg[diff > 0] = diff[ diff>0 ]
    down_chg[diff < 0] = diff[ diff < 0 ]
    up_chg_avg   = up_chg.ewm(com=time_window-1 , min_periods=time_window).mean()
    down_chg_avg = down_chg.ewm(com=time_window-1 , min_periods=time_window).mean()
    rs = abs(up_chg_avg/down_chg_avg)
    rsi = 100 - 100/(1+rs)
    return rsi
amzn['RSI'] = RSI(amzn['Close'], 14)
amzn['RSI70'] = 70.0
amzn['RSI50'] = 50.0
# custom indicator
def RSIaverage(data, n1, n2):
    RSI_1 = RSI(data, n1)
    RSI_2 = RSI(data, n2)
    return (RSI_1 + RSI_2) / 2
amzn['RSIaverage'] = RSIaverage(amzn['Close'], 2, 14)
# data ranges
piece_d = pd.date_range(start='2017-10-30', end='2020-10-01')
#piece_d = pd.date_range(start='2020-01-01', end='2020-10-01')
amzn_piece_d = amzn.reindex(piece_d)
data = pd.DataFrame(amzn_piece_d.dropna())
# sample of divergence signal
divergence = [[('2020-07-22',3250),('2020-09-01',3550)]]
data['divergence'] = np.where((data.index == '2020-07-10') | (data.index == '2020-09-01'), data['RSI'], None)
data['divergence'] = data['divergence'].dropna()

Requirement already satisfied: pandas_datareader in /opt/conda/lib/python3.8/site-packages (0.9.0)
Requirement already satisfied: pandas>=0.23 in /opt/conda/lib/python3.8/site-packages (from pandas_datareader) (1.1.5)
Requirement already satisfied: requests>=2.19.0 in /opt/conda/lib/python3.8/site-packages (from pandas_datareader) (2.25.1)
Requirement already satisfied: lxml in /opt/conda/lib/python3.8/site-packages (from pandas_datareader) (4.6.2)
Requirement already satisfied: python-dateutil>=2.7.3 in /opt/conda/lib/python3.8/site-packages (from pandas>=0.23->pandas_datareader) (2.8.1)
Requirement already satisfied: pytz>=2017.2 in /opt/conda/lib/python3.8/site-packages (from pandas>=0.23->pandas_datareader) (2020.5)
Requirement already satisfied: numpy>=1.15.4 in /opt/conda/lib/python3.8/site-packages (from pandas>=0.23->pandas_datareader) (1.19.4)
Requirement already satisfied: six>=1.5 in /opt/conda/lib/python3.8/site-packages (from python-dateutil>=2.7.3->pandas>=0.23->pandas_datareader) (1.15.0)
Requirement already satisfied: urllib3<1.27,>=1.21.1 in /opt/conda/lib/python3.8/site-packages (from requests>=2.19.0->pandas_datareader) (1.26.2)
Requirement already satisfied: certifi>=2017.4.17 in /opt/conda/lib/python3.8/site-packages (from requests>=2.19.0->pandas_datareader) (2020.12.5)
Requirement already satisfied: idna<3,>=2.5 in /opt/conda/lib/python3.8/site-packages (from requests>=2.19.0->pandas_datareader) (2.10)
Requirement already satisfied: chardet<5,>=3.0.2 in /opt/conda/lib/python3.8/site-packages (from requests>=2.19.0->pandas_datareader) (4.0.0)

You can have to see your strategy and the plot is your solution. If you have not defined which indicators you want to use, you can see them all together.

[2]:

!pip install mplfinance
# plot with candle daily, sma 20, sma 50, bb, RSI and yahoo style
import mplfinance as mpf
kwargs = dict(type='candle',volume=True,figratio=(16,9),figscale=2)
aps = [
    mpf.make_addplot(data['SMA20'],color='C0'), # blue
    mpf.make_addplot(data['SMA50'],color='C1'), # orange
#    mpf.make_addplot(data['EMA20'],color='C0'), # blue
#    mpf.make_addplot(data['EMA50'],color='C1'), # orange
    mpf.make_addplot(data['Upper Band'],linestyle='-.',color='g'),
    mpf.make_addplot(data['Lower Band'],linestyle='-.',color='g'),
    mpf.make_addplot(data['RSI'],color='C4',panel=2,ylabel='RSI'),
    mpf.make_addplot(data['RSI70'],color='g',panel=2,type='line',linestyle='-.',alpha=0.5),
    mpf.make_addplot(data['RSI50'],color='r',panel=2,type='line',linestyle='-.',alpha=0.5),
    mpf.make_addplot(data['divergence'],color='C0',panel=2,type='scatter',markersize=78,marker='o'),
    mpf.make_addplot(data['RSIaverage'],color='C2',panel=3,ylabel='RSIaverage'),
]
mpf.plot(data,**kwargs,style='yahoo',title='AMZN',addplot=aps,alines=divergence,fill_between=dict(y1=data['Lower Band'].values,y2=data['Upper Band'].values,alpha=0.1,color='g'))
#mpf.plot(data,**kwargs,style='yahoo',title='AMZN',addplot=aps,alines=divergence,fill_between=dict(y1=data['Lower Band'].values,y2=data['Upper Band'].values,alpha=0.1,color='g'),savefig=dict(fname='plot.with.candle.daily.sma.20.sma.50.bb.RSI.yahoo.style.png'))

Requirement already satisfied: mplfinance in /opt/conda/lib/python3.8/site-packages (0.12.7a4)
Requirement already satisfied: matplotlib in /opt/conda/lib/python3.8/site-packages (from mplfinance) (3.3.3)
Requirement already satisfied: pandas in /opt/conda/lib/python3.8/site-packages (from mplfinance) (1.1.5)
Requirement already satisfied: pyparsing!=2.0.4,!=2.1.2,!=2.1.6,>=2.0.3 in /opt/conda/lib/python3.8/site-packages (from matplotlib->mplfinance) (2.4.7)
Requirement already satisfied: pillow>=6.2.0 in /opt/conda/lib/python3.8/site-packages (from matplotlib->mplfinance) (8.0.1)
Requirement already satisfied: numpy>=1.15 in /opt/conda/lib/python3.8/site-packages (from matplotlib->mplfinance) (1.19.4)
Requirement already satisfied: python-dateutil>=2.1 in /opt/conda/lib/python3.8/site-packages (from matplotlib->mplfinance) (2.8.1)
Requirement already satisfied: cycler>=0.10 in /opt/conda/lib/python3.8/site-packages (from matplotlib->mplfinance) (0.10.0)
Requirement already satisfied: kiwisolver>=1.0.1 in /opt/conda/lib/python3.8/site-packages (from matplotlib->mplfinance) (1.3.1)
Requirement already satisfied: six in /opt/conda/lib/python3.8/site-packages (from cycler>=0.10->matplotlib->mplfinance) (1.15.0)
Requirement already satisfied: pytz>=2017.2 in /opt/conda/lib/python3.8/site-packages (from pandas->mplfinance) (2020.5)
_images/python_strategies_3_1.png

How to load signals on your strategy

The best practice is to prepare one column for each signal that you want to use on your strategy. The strategy below use the moving averages that they are SMA and EMA. You can use one or the other.

Disclaimer

The strategies below are some simple samples for having an idea how to use the libraries: those strategies are for the educational purpose only. All investments and trading in the stock market involve risk: any decisions related to buying/selling of stocks or other financial instruments should only be made after a thorough research, backtesting, running in demo and seeking a professional assistance if required.

Moving Average Crossover Strategy - Sample 1

  • when the price value crosses the MA value from below, it will close any existing short position and go long (buy) one unit of the asset

  • when the price value crosses the MA value from above, it will close any existing long position and go short (sell) one unit of the asset

Reference: https://www.learndatasci.com/tutorials/python-finance-part-3-moving-average-trading-strategy/

[3]:

# Moving Average Crossover Strategy - Sample 1
#ma = 'SMA20'
ma = 'EMA20'
# Taking the difference between the prices and the MA timeseries
data['price_ma_diff'] = data['Close'] - data[ma]
# Taking the sign of the difference to determine whether the price or the EMA is greater
data['signal1'] = data['price_ma_diff'].apply(np.sign)
data['position1'] = data['signal1'].diff()
data['buy'] = np.where(data['position1'] == 2, data[ma], np.nan)
data['sell'] = np.where(data['position1'] == -2, data[ma], np.nan)

[4]:

# plot with candle daily, sma 20, signal and yahoo style
import mplfinance as mpf
kwargs = dict(type='candle',figratio=(16,9),figscale=2)
aps = [
    mpf.make_addplot(data[ma],color='C0'), # blue
    mpf.make_addplot(data['buy'],color='g',type='scatter',markersize=78,marker='^'),
    mpf.make_addplot(data['sell'],color='r',type='scatter',markersize=78,marker='v'),
    mpf.make_addplot(data['signal1'],color='C1',panel=1,ylabel='signal')
]
mpf.plot(data,**kwargs,style='yahoo',title='AMZN',addplot=aps)
#mpf.plot(data,**kwargs,style='yahoo',title='AMZN',addplot=aps,savefig=dict(fname='plot.with.candle.daily.sma.20.signal.yahoo.style.png'))
_images/python_strategies_6_0.png 
[5]:

!pip install backtesting
from backtesting import Backtest, Strategy
from backtesting.lib import crossover
from backtesting.test import SMA
class PositionSign(Strategy):
    def init(self):
        self.close = self.data.Close
    def next(self):
        i = len(self.data.Close) - 1
        if self.data.position1[-1] == 2:
            self.position.close()
            self.buy()
        elif self.data.position1[-1] == -2:
            self.position.close()
            self.sell()
bt = Backtest(data, PositionSign, cash=10000, commission=.002, exclusive_orders=True)
bt.run()

Requirement already satisfied: backtesting in /opt/conda/lib/python3.8/site-packages (0.3.0)
Requirement already satisfied: bokeh>=1.4.0 in /opt/conda/lib/python3.8/site-packages (from backtesting) (2.2.3)
Requirement already satisfied: numpy in /opt/conda/lib/python3.8/site-packages (from backtesting) (1.19.4)
Requirement already satisfied: pandas!=0.25.0,>=0.25.0 in /opt/conda/lib/python3.8/site-packages (from backtesting) (1.1.5)
Requirement already satisfied: python-dateutil>=2.1 in /opt/conda/lib/python3.8/site-packages (from bokeh>=1.4.0->backtesting) (2.8.1)
Requirement already satisfied: Jinja2>=2.7 in /opt/conda/lib/python3.8/site-packages (from bokeh>=1.4.0->backtesting) (2.11.2)
Requirement already satisfied: packaging>=16.8 in /opt/conda/lib/python3.8/site-packages (from bokeh>=1.4.0->backtesting) (20.8)
Requirement already satisfied: pillow>=7.1.0 in /opt/conda/lib/python3.8/site-packages (from bokeh>=1.4.0->backtesting) (8.0.1)
Requirement already satisfied: PyYAML>=3.10 in /opt/conda/lib/python3.8/site-packages (from bokeh>=1.4.0->backtesting) (5.3.1)
Requirement already satisfied: typing-extensions>=3.7.4 in /opt/conda/lib/python3.8/site-packages (from bokeh>=1.4.0->backtesting) (3.7.4.3)
Requirement already satisfied: tornado>=5.1 in /opt/conda/lib/python3.8/site-packages (from bokeh>=1.4.0->backtesting) (6.1)
Requirement already satisfied: MarkupSafe>=0.23 in /opt/conda/lib/python3.8/site-packages (from Jinja2>=2.7->bokeh>=1.4.0->backtesting) (1.1.1)
Requirement already satisfied: pyparsing>=2.0.2 in /opt/conda/lib/python3.8/site-packages (from packaging>=16.8->bokeh>=1.4.0->backtesting) (2.4.7)
Requirement already satisfied: pytz>=2017.2 in /opt/conda/lib/python3.8/site-packages (from pandas!=0.25.0,>=0.25.0->backtesting) (2020.5)
Requirement already satisfied: six>=1.5 in /opt/conda/lib/python3.8/site-packages (from python-dateutil>=2.1->bokeh>=1.4.0->backtesting) (1.15.0)

/opt/conda/lib/python3.8/site-packages/backtesting/_plotting.py:45: UserWarning: Jupyter Notebook detected. Setting Bokeh output to notebook. This may not work in Jupyter clients without JavaScript support (e.g. PyCharm, Spyder IDE). Reset with `backtesting.set_bokeh_output(notebook=False)`.
  warnings.warn('Jupyter Notebook detected. '
Loading BokehJS ... 
[5]:

Start                     2018-01-10 00:00:00
End                       2020-10-01 00:00:00
Duration                    995 days 00:00:00
Exposure Time [%]                     96.9432
Equity Final [$]                      6449.39
Equity Peak [$]                       11361.5
Return [%]                           -35.5061
Buy & Hold Return [%]                 156.811
Return (Ann.) [%]                    -14.8609
Volatility (Ann.) [%]                 25.1543
Sharpe Ratio                                0
Sortino Ratio                               0
Calmar Ratio                                0
Max. Drawdown [%]                    -62.3199
Avg. Drawdown [%]                    -17.0591
Max. Drawdown Duration      934 days 00:00:00
Avg. Drawdown Duration      238 days 00:00:00
# Trades                                   68
Win Rate [%]                               25
Best Trade [%]                        58.9442
Worst Trade [%]                       -12.944
Avg. Trade [%]                      -0.750267
Max. Trade Duration         141 days 00:00:00
Avg. Trade Duration          15 days 00:00:00
Profit Factor                        0.790866
Expectancy [%]                      -0.473583
SQN                                  -1.05025
_strategy                        PositionSign
_equity_curve                             ...
_trades                       Size  EntryB...
dtype: object

[6]:

bt.plot()
Moving Average Crossover Strategy - Sample 2

  • when the short term moving average crosses above the long term moving average, this indicates a buy signal

  • when the short term moving average crosses below the long term moving average, it may be a good moment to sell

Reference: https://towardsdatascience.com/making-a-trade-call-using-simple-moving-average-sma-crossover-strategy-python-implementation-29963326da7a

[7]:

# Moving Average Crossover Strategy - Sample 2
ma20 = 'SMA20'
ma50 = 'SMA50'
#ma20 = 'EMA20'
#ma50 = 'EMA50'
data['signal2'] = 0.0
data['signal2'] = np.where(data[ma20] > data[ma50], 1.0, 0.0)
data['position2'] = data['signal2'].diff()
data['buy'] = np.where(data['position2'] == 1, data[ma20], np.nan)
data['sell'] = np.where(data['position2'] == -1, data[ma20], np.nan)

[8]:

# plot with candle daily, sma 20, sma 50, signal and yahoo style
import mplfinance as mpf
kwargs = dict(type='candle',figratio=(16,9),figscale=2)
aps = [
    mpf.make_addplot(data[ma20],color='C0'), # blue
    mpf.make_addplot(data[ma50],color='C1'), # orange
    mpf.make_addplot(data['buy'],color='g',type='scatter',markersize=78,marker='^'),
    mpf.make_addplot(data['sell'],color='r',type='scatter',markersize=78,marker='v'),
    mpf.make_addplot(data['signal2'],color='C1',panel=1,ylabel='signal')
]
mpf.plot(data,**kwargs,style='yahoo',title='AMZN',addplot=aps)
#mpf.plot(data,**kwargs,style='yahoo',title='AMZN',addplot=aps,savefig=dict(fname='plot.with.candle.daily.sma.20.sma.50.signal.yahoo.style.png'))
_images/python_strategies_11_0.png 
[9]:

from backtesting import Backtest, Strategy
from backtesting.lib import crossover
from backtesting.test import SMA
class SmaCross(Strategy):
    n1 = 20
    n2 = 50
    def init(self):
        close = self.data.Close
        self.sma1 = self.I(SMA, close, self.n1)
        self.sma2 = self.I(SMA, close, self.n2)
    def next(self):
        if crossover(self.sma1, self.sma2):
            self.position.close()
            self.buy()
        elif crossover(self.sma2, self.sma1):
            self.position.close()
            self.sell()
bt = Backtest(data, SmaCross, cash=10000, commission=.002, exclusive_orders=True)
bt.run()

[9]:

Start                     2018-01-10 00:00:00
End                       2020-10-01 00:00:00
Duration                    995 days 00:00:00
Exposure Time [%]                     90.5386
Equity Final [$]                      7712.69
Equity Peak [$]                       12532.8
Return [%]                           -22.8731
Buy & Hold Return [%]                 156.811
Return (Ann.) [%]                    -9.08703
Volatility (Ann.) [%]                  24.431
Sharpe Ratio                                0
Sortino Ratio                               0
Calmar Ratio                                0
Max. Drawdown [%]                    -52.0307
Avg. Drawdown [%]                    -11.5153
Max. Drawdown Duration      517 days 00:00:00
Avg. Drawdown Duration       80 days 00:00:00
# Trades                                   13
Win Rate [%]                          38.4615
Best Trade [%]                        35.0832
Worst Trade [%]                      -32.3988
Avg. Trade [%]                       -2.23977
Max. Trade Duration         166 days 00:00:00
Avg. Trade Duration          70 days 00:00:00
Profit Factor                         0.79924
Expectancy [%]                       -1.11657
SQN                                 -0.620588
_strategy                            SmaCross
_equity_curve                             ...
_trades                       Size  EntryB...
dtype: object

[10]:

bt.plot()

Pine

Prerequisites

Tradingview is an analysis platform where you can,

  • analyze your trades by indicators, your scripts and perform your backtesting

  • share your analysis with other traders and investors, because it is a full-fledged social network

  • view the analysis of other users

  • monitor your trades by alerts

For implementing your indicators or your strategies, Tradingview has own script language: Pine.

For using Pine,

  • just access the Chart section

  • and click on the tab named Pine Editor on the bottom

  • it is there that you can copy your Pine scripts

You can find all scripts descripted in this tutorial on GitHub in the folder src/pine/.

Getting started

Pine is a script language own of Tradingview for adding your indicators and backtesting your trades.

This is a summary of the pine language syntax: you can find,

Basics

There are some methods to load your data

//@version=4
//study("My Simple Sample Script") // in a chart space dedicated
//study("My Simple Sample Script", max_labels_count = 500) // in a chart space dedicated with a max label > 50, that it is the default
study("My Simple Sample Script", "", true) // in the main chart

It is important to remember that it is a script language:

Pine draws your indicators and data in the Tradingview charts and in the Data Window. So, it is important to understand which methods you have for debugging your code.

// initialization of a variable
my_numeric_variable = 1
my_string_variable = "H"

// where print something
plotchar(my_numeric_variable, "Numeric variable", "") // on the right, in Data Window
plotchar(bar_index, "Bar Index", "i", location = location.top) // on the right, in Data Window and on the main panel, in the chart space
label.new(bar_index, high, my_string_variable) // on each bar
if barstate.islast
    label.new(bar_index, low, "L", style = label.style_label_up) // only on the last bar
https://s3-eu-west-1.amazonaws.com/cdn.bilardi.net/backtesting/tool-comparison/pine/getting-started-basics.png

It is not possible to use

  • plotchar method on Data Window with a string value

  • plotchar and label.new methods in a function

  • plotchar method in a local scope like also a for loop

Arrays

Pine arrays are one-dimensional. All elements of an array are of the same type, which can be int, float, bool or color, always of series form.

// initialization of a vector
my_numeric_array = array.new_float(0)
for i = 1 to 6
    array.push(my_numeric_array, i)
if barstate.islast // print
    label.new(bar_index, high, "my_numeric_array:\n" + tostring(my_numeric_array), style = label.style_label_down, size = size.large)

my_string_array = array.new_string(1, "hello")
array.concat(my_string_array, array.new_string(1, "world"))
array.concat(my_string_array, array.new_string(1, "!"))
if barstate.islast // print
    label.new(bar_index, low, "my_string_array: " + array.get(my_string_array, 0), style = label.style_label_up, size = size.large)

my_logic_array = array.new_bool(0)
array.push(my_logic_array, 0)
array.push(my_logic_array, 1)

// operations with arrays
sum = array.sum(my_numeric_array)
mean = array.avg(my_numeric_array)
median = array.median(my_numeric_array)
plotchar(sum, "sum", "")
https://s3-eu-west-1.amazonaws.com/cdn.bilardi.net/backtesting/tool-comparison/pine/getting-started-arrays.png

Weighted average

performance = array.new_float(0)
array.push(performance, 3)
array.push(performance, 7)
array.push(performance,11)
weight = array.new_float(0)
array.push(weight, .3)
array.push(weight, .4)
array.push(weight, .3)
weighted_average = array.get(performance, 0) * array.get(weight, 0) + array.get(performance, 1) * array.get(weight, 1) + array.get(performance, 2) * array.get(weight, 2)
plotchar(weighted_average, "weighted_average", "")
https://s3-eu-west-1.amazonaws.com/cdn.bilardi.net/backtesting/tool-comparison/pine/getting-started-wa.png

Functions

The functions could return something or not (see the docs for details). The first function named f_print, it returns nothing, only prints on the chart.

// prints on last bar, in the chart space
f_print(_text) =>
    // Create label on the first bar.
    var _label = label.new(bar_index, na, _text, xloc.bar_index, yloc.price, color(na), label.style_none, color.gray, size.large, text.align_left)
    // On next bars, update the label's x and y position, and the text it displays.
    label.set_xy(_label, bar_index, highest(10)[1])
    label.set_text(_label, _text)
f_print("My details:\nHigh = " + tostring(high))
f_print(my_string_variable + "ello world!\n\n\n")
https://s3-eu-west-1.amazonaws.com/cdn.bilardi.net/backtesting/tool-comparison/pine/getting-started-labels.png

The second function named multiply_performance_weight, it returns an array.

// portfolio performance
company = array.new_string(1, "AAPL")
array.concat(company, array.new_string(1, "NFLX"))
array.concat(company, array.new_string(1, "AMZN"))
multiply_performance_weight(_performance, _weight) =>
    _performance_weight = array.new_float(0)
    for i = 0 to array.size(_performance) - 1
        _one_of_performance_weight = array.get(_performance, i) * array.get(_weight, i)
        array.push(_performance_weight, _one_of_performance_weight)
    _performance_weight // return the array

And inner a for loop is not possible to use plotchar method, but only label.new method.

performance_weight = multiply_performance_weight(performance, weight)
weighted_average2 = array.sum(performance_weight)
for i = 0 to array.size(performance) - 1
    if barstate.islast // print
        label.new(bar_index, close, array.get(company, i) + ": " + tostring(array.get(performance_weight, i)), style = label.style_label_up, size = size.large)
plotchar(weighted_average2, "portfolio", "") // Cannot use 'plotchar' in local scope.
https://s3-eu-west-1.amazonaws.com/cdn.bilardi.net/backtesting/tool-comparison/pine/getting-started-portfolio.png

Historical data

Tradingview displays directly the data in the main chart panel, and you can see the all data moving the main chart back.

So, this step is not necessary for using the Pine script language because it is already integrated.

The unique limit is that you can only use the assets present in Tradingview and its partners features.

https://s3-eu-west-1.amazonaws.com/cdn.bilardi.net/backtesting/tool-comparison/pine/historical-data.png

Graphics

This is a sample for drawing objects with Pine. There are 2 principal methods for drawing, you can draw on

Lines

//@version=4
study("My Simple Sample Script", "", true) // the main chart

You can use the arrays pre loaded for you, like close (values of the closes), and modify them loading only the values that you want

// monthly points - the first Monday of each month
custom = array.new_float(1)
if dayofmonth <= 7 and dayofweek == dayofweek.monday
    array.set(custom, 0, close)
else
    array.set(custom, 0, na)
customPreviousClose = array.get(custom, 0)[1]
plot(customPreviousClose, "customPreviousClose", color.yellow, 6, plot.style_circles)
plot(customPreviousClose, "customPreviousClose", color.yellow, 6, plot.style_line, transp = 50)

And you can add many many indicators in a few code lines

// moving average
sma25 = sma(close, 25)
ema25 = ema(close, 25)
plot(sma25, "sma 25", color.orange, 2)
plot(ema25, "ema 25", color.red, 2)

// bollinger bands
ma = sma25
dev = 2 * stdev(close, 25)
upper = ma + dev
lower = ma - dev
p1 = plot(upper, color=color.green, linewidth=1, style=plot.style_circles)
p2 = plot(lower, color=color.green, linewidth=1, style=plot.style_circles)
fill(p1, p2)
https://s3-eu-west-1.amazonaws.com/cdn.bilardi.net/backtesting/tool-comparison/pine/graphics-lines.png

It is possible to draw line from one point to another point, by bar_index or time

// line by index
i = line.new(bar_index, high, bar_index[10], low[10], width = 4)
line.delete(i[1])

// line by time
t = line.new(time, high, time[10], low[10], xloc.bar_time, width = 4)
line.delete(t[1])
https://s3-eu-west-1.amazonaws.com/cdn.bilardi.net/backtesting/tool-comparison/pine/graphics-lines-delete.png

Columns

It is possible to simulate Volume indicator on the main chart with specific parameters

  • on the study method

  • adding a workaround that it is the plot of high line

//@version=4
study("My Simple Volume Script", "", true, format.volume, 0, scale.none)

//Get volume for current bar and multiply with vwap
vInverse = volume * vwap
plot(series=vInverse, title="Volume", style=plot.style_columns, color=close >= open ? color.green : color.red, transp=50)

//Plot high line to scale down the columns
highLine = input(1000, "High Line", minval = 2, step = 100)
limit = highest(vInverse, highLine)
scaleFactor = 100 / input(20, "% of vertical space used", step = 10, maxval = 100)
plot(limit * scaleFactor, "Historical High", color.black)
https://s3-eu-west-1.amazonaws.com/cdn.bilardi.net/backtesting/tool-comparison/pine/graphics-columns.png

And you can also add SMA indicator over the Volume indicator

//Make the moving average user configurable
showMA = input(true)
plot(showMA ? sma(vInverse,20) : na, title="SMA", style=plot.style_area, color=color.white, transp=80)

The second script has also some inputs that you can manage the values by Settings symbol > Input.

https://s3-eu-west-1.amazonaws.com/cdn.bilardi.net/backtesting/tool-comparison/pine/graphics-columns-show-ma-true.png

And you can change all colors by Settings symbol > Style.

Points

There are many ready-made methods for you like pivot points.

//@version=4
study("My Simple Pivot Script", "", true)

In that sample you can decide if which plots show by input methods.

legs = input(14)
showPivotHigh = input(true)
showPivotLow = input(true)

And it is used a function, for reusing the code in two points.

pivot_points(pi, legs) =>
    newPi = not na(pi)
    var float ci = na
    if newPi
        ci := close[legs]
    [pi, newPi, ci]

[pHi, newPHi, cHi] = pivot_points(pivothigh(close, legs, legs), legs)
plot(showPivotHigh ? cHi : na, "", newPHi ? na : color.fuchsia, offset = - legs)
plotchar(showPivotHigh ? pHi : na, "pHi", "▼", location.top, color.fuchsia, offset = - legs)

[pLi, newPLi, cLi] = pivot_points(pivotlow(close, legs, legs), legs)
plot(showPivotLow ? cLi: na, "", newPLi ? na : color.orange, offset = - legs)
plotchar(showPivotLow ? pLi: na, "pLi", "▲", location.bottom, color.orange, offset = - legs)
https://s3-eu-west-1.amazonaws.com/cdn.bilardi.net/backtesting/tool-comparison/pine/graphics-pivot.png

Strategies

You can initilize a strategy by Pine language.

Tradingview has a tab named Strategy Tester near the tab named Pine Editor that it has been created for backtesting your Pine scripts.

How to load indicators on your strategy

When you want to test your strategy on Tradingview, you have to start your Pine script with the method strategy.

//@version=4
strategy("My Simple Strategy Script", "", true) // in the main chart

And then, it is the same for loading your indicators.

A script runs either in overlay=true mode on the chart, in which case it cannot direct plots elsewhere, or in a separate pane when overlay=false (the default).

But there is a workaround for having some plots in the main chart and others in another chart. You can use Settings > Style for flagging who see where you want. So you can also add this code,

// initialization
legs = input(20)

// moving averages
sma20 = sma(close, legs)
plot(sma20, "sma 20", color.blue, 2)
plot(sma(close, 50), "sma 50", color.orange, 2)

// bollinger bands
dev = 2 * stdev(close, legs)
upper = sma20 + dev
lower = sma20 - dev
p1 = plot(upper, title="BB upper", color=color.green, linewidth=1, style=plot.style_circles)
p2 = plot(lower, title="BB lower", color=color.green, linewidth=1, style=plot.style_circles)
fill(p1, p2)

// volume
plot(volume * vwap, title="Volume", style=plot.style_columns, color=close >= open ? color.green : color.red, transp=50)

// rsi
plot(rsi(close, legs), title="RSI", color=color.purple)
h0 = hline(70)
h1 = hline(50)
fill(h0, h1, color=color.purple, transp=75)

Add the chart on the main chart with the header

//@version=4
strategy("My Simple Strategy Script", "", true) // in the main chart

Add the chart on another chart with the header

//@version=4
strategy("My Simple Strategy Script") // a chart space dedicated

And then you can use the Settings for splitting the plots between the main chart and others.

https://s3-eu-west-1.amazonaws.com/cdn.bilardi.net/backtesting/tool-comparison/pine/strategies-indicators.png

Instead, for the line method it is necessary to use an input and conditions for splitting the lines between the main chart and another. You can use Settings > Input for flagging who see where you want in order of the conditions.

In the code below is used the variable named showLineRSI for the visualization of the lines on RSI chart or the main chart.

// initialization
showLineRSI = input(true)

var bar_time1 = 0
var bar_time2 = 0
var float bar_high1 = 0
var float bar_high2 = 0
var float bar_rsi1 = 0
var float bar_rsi2 = 0
_rsi = rsi(close, 20)

if time == timestamp(2020, 04, 16, 09, 30)
    bar_high1 := high
    bar_time1 := time
    bar_rsi1 := _rsi

if time == timestamp(2020, 05, 21, 09, 30)
    bar_high2 := high
    bar_time2 := time
    bar_rsi2 := _rsi

// rsi line
plot(bar_rsi1 == 0 or bar_rsi2 != 0 ? na : _rsi, title="RSI of trend", color=color.orange, linewidth=2)
plot(rsi(close, 20), title="RSI", color=color.purple)
h0 = hline(70)
h1 = hline(50)
fill(h0, h1, color=color.purple, transp=75)

// rsi trend line
if bar_rsi1 != 0 and bar_rsi2 != 0 and showLineRSI
    line.new(bar_time2, bar_rsi2, bar_time1, bar_rsi1, xloc.bar_time, width = 4)
    bar_rsi1 := 0
    bar_rsi2 := 0

// high line
plot(bar_high1 == 0 or bar_high2 != 0 ? na : high, title="High line", color=color.orange, linewidth=6)

// high trend line
if bar_high1 != 0 and bar_high2 != 0 and showLineRSI != true
    line.new(bar_time2, bar_high2, bar_time1, bar_high1, xloc.bar_time, width = 4)
    bar_high1 := 0
    bar_high2 := 0

And then you can use the Settings and,

  • Style, for splitting the plots between the main chart and others

  • Input, for splitting the lines between the main chart and others

In the main chart, for selecting the plots, you can use Settings > Style

https://s3-eu-west-1.amazonaws.com/cdn.bilardi.net/backtesting/tool-comparison/pine/strategies-trend-lines-1.png

And Settings > Input, for showing the line on the main chart

https://s3-eu-west-1.amazonaws.com/cdn.bilardi.net/backtesting/tool-comparison/pine/strategies-trend-lines-2.png
https://s3-eu-west-1.amazonaws.com/cdn.bilardi.net/backtesting/tool-comparison/pine/strategies-trend-lines-3.png

Then, in the other chart, for selecting the plots, you can use Settings > Style

https://s3-eu-west-1.amazonaws.com/cdn.bilardi.net/backtesting/tool-comparison/pine/strategies-trend-lines-4.png

How to load signals on your strategy

The best practice is to prepare one variable for each signal that you want to use on your strategy. The strategy below use the moving averages that they are SMA and EMA. You can use one or the other.

Disclaimer

The strategies below are some simple samples for having an idea how to use the libraries: those strategies are for the educational purpose only. All investments and trading in the stock market involve risk: any decisions related to buying/selling of stocks or other financial instruments should only be made after a thorough research, backtesting, running in demo and seeking a professional assistance if required.

Moving Average Crossover Strategy - Sample 1

  • when the price value crosses the MA value from below, it will close any existing short position and go long (buy) one unit of the asset

  • when the price value crosses the MA value from above, it will close any existing long position and go short (sell) one unit of the asset

Reference: https://www.learndatasci.com/tutorials/python-finance-part-3-moving-average-trading-strategy/

//@version=4
//strategy("My Simple Strategy Script - sample 1") // a chart space dedicated
strategy("My Simple Strategy Script - sample 1", "", true) // in the main chart

legs = input(20)
//ma = sma(close, legs)
ma = ema(close, legs)
plot(ma, "Ma", color.orange, 2)
dev = 2 * stdev(close, legs)
upper = ma + dev
lower = ma - dev
p1 = plot(upper, title="BB upper", color=color.green, linewidth=1, style=plot.style_circles)
p2 = plot(lower, title="BB lower", color=color.green, linewidth=1, style=plot.style_circles)
fill(p1, p2)

// Taking the difference between the prices and the MA timeseries
price_ma_diff = close - ma

// Taking the sign of the difference to determine whether the price or the EMA is greater
position = price_ma_diff - price_ma_diff[1]
buyEntry = position >= 2 ? price_ma_diff : na
sellEntry = position <= -2 ? price_ma_diff : na

if (buyEntry)
//    strategy.entry("MaLE", strategy.long, stop=lower, oca_name="Ma", oca_type=strategy.oca.cancel, comment="MaLE")
    strategy.entry("MaLE", strategy.long, oca_name="Ma", oca_type=strategy.oca.cancel, comment="MaLE")
else
    strategy.cancel(id="MaLE")

if (sellEntry)
//    strategy.entry("MaSE", strategy.short, stop=upper, oca_name="Ma", oca_type=strategy.oca.cancel, comment="MaSE")
    strategy.entry("MaSE", strategy.short, oca_name="Ma", oca_type=strategy.oca.cancel, comment="MaSE")
else
    strategy.cancel(id="MaSE")

plot(strategy.equity, title="Equity", color=color.red, linewidth=2, style=plot.style_areabr)
https://s3-eu-west-1.amazonaws.com/cdn.bilardi.net/backtesting/tool-comparison/pine/strategies-sample1.png

When you Add to Chart your strategy, the Strategy Tester tab will contain the results of your backtesting:

  • an Overview of the main parameters like Net Profit and Drawdown

https://s3-eu-west-1.amazonaws.com/cdn.bilardi.net/backtesting/tool-comparison/pine/strategies-sample1-overview.png

  • the Performance Summary with all important parameters for defining if your backtesting is good or not

https://s3-eu-west-1.amazonaws.com/cdn.bilardi.net/backtesting/tool-comparison/pine/strategies-sample1-summary.png

  • the List of Trades with the details of each trade like Date, Price and Profit

https://s3-eu-west-1.amazonaws.com/cdn.bilardi.net/backtesting/tool-comparison/pine/strategies-sample1-trades.png
Moving Average Crossover Strategy - Sample 2

  • when the short term moving average crosses above the long term moving average, this indicates a buy signal

  • when the short term moving average crosses below the long term moving average, it may be a good moment to sell

Reference: https://towardsdatascience.com/making-a-trade-call-using-simple-moving-average-sma-crossover-strategy-python-implementation-29963326da7a

//@version=4
//strategy("My Simple Strategy Script - sample 2") // a chart space dedicated
strategy("My Simple Strategy Script - sample 2", "", true) // in the main chart

ma20 = sma(close, 20)
ma50 = sma(close, 50)
//ma20 = ema(close, 20)
//ma50 = ema(close, 50)
plot(ma20, "Ma20", color.blue, 2)
plot(ma50, "Ma50", color.orange, 2)

buyEntry = crossover(ma20, ma50)
sellEntry = crossunder(ma20, ma50)

if (buyEntry)
    strategy.entry("MaLE", strategy.long, oca_name="Ma", oca_type=strategy.oca.cancel, comment="MaLE")
else
    strategy.cancel(id="MaLE")

if (sellEntry)
    strategy.entry("MaSE", strategy.short, oca_name="Ma", oca_type=strategy.oca.cancel, comment="MaSE")
else
    strategy.cancel(id="MaSE")

plot(strategy.equity, title="Equity", color=color.red, linewidth=2, style=plot.style_areabr)
https://s3-eu-west-1.amazonaws.com/cdn.bilardi.net/backtesting/tool-comparison/pine/strategies-sample2.png

When you Add to Chart your strategy, the Strategy Tester tab will contain the results of your backtesting:

  • Overview

https://s3-eu-west-1.amazonaws.com/cdn.bilardi.net/backtesting/tool-comparison/pine/strategies-sample2-overview.png

  • Performance Summary

https://s3-eu-west-1.amazonaws.com/cdn.bilardi.net/backtesting/tool-comparison/pine/strategies-sample2-summary.png

  • List of Trades

https://s3-eu-west-1.amazonaws.com/cdn.bilardi.net/backtesting/tool-comparison/pine/strategies-sample2-trades.png

Comparison

The comparison is for evaluating which language or system you could use based on:

  • maintenance of packages

  • learning curve / costs

  • performance

  • monitoring

In this tutorial, there have been used some languages and systems:

You could know another language or system that it is perfect for your needs. This is only a discussion to highlight key points, bottlenecks and risks.

Maintenance of packages

Learning curve / costs

Performance

Monitoring

Conclusion

Historical data

The samples that you find in this tutorial they use the same data reader: Yahoo Finance API.

This is a tutorial, so it is for educational purpose, and it is not important the results. Each historical data contains some differences among others, so it is very important understanding when using which source.

At the beginning, when you test your code, or learn your instruments, you can also use free historical data. In a perfect system, when your QT system will become your demo and your business, you have to use paid historical data of the trading platform where you will do your demo and your business.

Why do you have to use the same historical data of the trading platform where you will do your demo and your business ?

  • for avoiding the closing differences or other, that they would completely change the backtesting results

  • for being prepared for any issue related to those data that will be used for your business (see the Comparison section for details)

Analysis

You cannot make backtesting without defining your trading strategy. And you may not use your trading strategy live without analyze the financial historical data by backtesting.

You can use existing indicators and yours for defining resistence points and signals.

Coming soon.

Best practice

You can use the command line, an IDE or a (Jupyter) notebook for your favorite language or directly a trading platform console, but the rules below do not change: you have to

  • evaluate your data source and to check (*) that your driver is working properly

  • test your code by TDD every time that you change it

  • check (*) your algorithm and its parameters by your backtesting system

  • evaluate the hosting where you run your QT system and to check (*) that your QT system did not reach its limits

  • define when your QT system does not have to work

(*) it would be great if each check can run to warn you with a channel you follow about

  • information of the next release of your driver, your hosting, .. or your trading platform

  • your algorithm performance

At the beginning, the checks can be executed with a wide cadence, until you find how often each check should be performed according to your expectations. In a perfect system, when your QT system will become your business, the checks will be executed continuously or with a tight cadence.

Coming soon # mql4/index