Table with predicted price change

Description

Table with predicted price change

Usage

DFR

Format

A dataframe with one column

"MagicNumber.x

Unique identifiers of the Trading Robots from Trade Log

TicketNumber

Ticket Number of closed position

OrderStartTime

Date and Time when order started

OrderCloseTime

Date and Time when order closed

Profit

Monetary result of the trade

Symbol

Symbol of the Asset e.g. EURUSD

OrderType

Order Type 0 - buy, 1 - sell

"CUMSUM_PNL

Cumulative sum of Profit and Loss

Table with indicator and price change dataset

Description

Table with indicator and price change dataset

Usage

EURUSDM15X75

Format

A dataframe with several columns

LABEL

future price change

X1-X75

Values of the macd indicator

Table with Trade States and sample of actual policy for those states

Description

Table with Trade States and sample of actual policy for those states

Usage

TradeStatePolicy

Format

A dataframe with 2 columns:

TradeState

Current trade state status

Policy

Policy choice

Function to read, transform, aggregate and save data for further retraining of regression model for a single asset

Description

Function is collecting data from the csv files Data objects are transformed to be suitable for Regression Modelling. Price change will be in the column 'LABEL', column X1 will keep the time index Result will be written to a new or aggregated to the existing '.rds' file

Function is keeping generated dataset to be not larger than specified by the user

Usage

aml_collect_data(
  indicator_dataset,
  symbol,
  timeframe = 60,
  path_data,
  max_nrows = 2500
)

Arguments

Details

Function is not handling shift of the price and indicator datasets.

This function is relying on the data collection from the dedicated data robot Other 'aml_*' functions will work based on the data processed by this function

Value

Function is writing files into Decision Support System folder, mainly file object with the model

Author(s)

(C) 2020, 2021 Vladimir Zhbanko

Examples

# write examples for the function
library(dplyr)
library(readr)
library(lubridate)
library(lazytrade)
library(magrittr)

# sample dataset
ind = system.file("extdata", "AI_RSIADXUSDJPY60.csv",
                  package = "lazytrade") %>% read_csv(col_names = FALSE)

# convert to POSIX format
ind$X1 <- ymd_hms(ind$X1)

# create temporary path (check output of tempdir() to check the result)
path_data <- normalizePath(tempdir(),winslash = "/")

# add tick data to the folder
tick = system.file("extdata", "TickSize_AI_RSIADX.csv",
                  package = "lazytrade") %>% read_csv(col_names = FALSE)

write_csv(tick, file.path(path_data, "TickSize_AI_RSIADX.csv"), col_names = FALSE)


# data transformation using the custom function for one symbol
aml_collect_data(indicator_dataset = ind,
                 symbol = 'USDJPY',
                 timeframe = 60,
                 path_data = path_data)

Function to consolidate model test results

Description

Function is designed to evaluate test results of multiple models. This is done to select only group of models with the best performance. In addition, function will provide facility to generate logs hence to allow tracking of long term model performance

Usage

aml_consolidate_results(
  timeframe = 15,
  used_symbols,
  path_model,
  path_sbxm,
  path_sbxs,
  min_quality = 0.75,
  get_quantile = FALSE,
  log_results = FALSE,
  path_logs = NULL
)

Arguments

Details

Provide a modular facility to aggregate and update files, write performance logs.

Value

Function is writing files into Decision Support System folders

Author(s)

(C) 2021 Vladimir Zhbanko

Examples

library(dplyr)
library(magrittr)
library(readr)
library(lazytrade)
library(stats)

testpath <- normalizePath(tempdir(),winslash = "/")
path_model <- file.path(testpath, "_Model")
path_sbxm <- file.path(testpath, "_T1")
path_sbxs <- file.path(testpath, "_T3")
path_logs <- file.path(testpath, "_LOGS")
dir.create(path_model)
dir.create(path_sbxm)
dir.create(path_sbxs)
dir.create(path_logs)

file.copy(from = system.file("extdata", "StrTest-EURGBPM15.csv", package = "lazytrade"),
          to = file.path(path_model, "StrTest-EURGBPM15.csv"), overwrite = TRUE)

file.copy(from = system.file("extdata", "StrTest-EURJPYM15.csv", package = "lazytrade"),
          to = file.path(path_model, "StrTest-EURJPYM15.csv"), overwrite = TRUE)

file.copy(from = system.file("extdata", "StrTest-EURUSDM15.csv", package = "lazytrade"),
          to = file.path(path_model, "StrTest-EURUSDM15.csv"), overwrite = TRUE)

Pairs <- c("EURGBP","EURJPY", "EURUSD")

aml_consolidate_results(timeframe = 15,
                        used_symbols = Pairs,
                        path_model = path_model,
                        path_sbxm = path_sbxm,
                        path_sbxs = path_sbxs,
                        min_quality = 0.75,
                        get_quantile = FALSE)



aml_consolidate_results(timeframe = 15,
                        used_symbols = Pairs,
                        path_model = path_model,
                        path_sbxm = path_sbxm,
                        path_sbxs = path_sbxs,
                        min_quality = 0.75,
                        get_quantile = FALSE,
                        log_results = TRUE,
                        path_logs = path_logs)

Function to train Deep Learning regression model for a single asset

Description

Function is training h2o deep learning model to match future prices of the asset to the indicator pattern. Main idea is to be able to predict future prices by solely relying on the recently retrieved indicator pattern. This is to mimic traditional algorithmic systems based on the indicator rule and to attempt automated optimization with AI.

Usage

aml_make_model(
  symbol,
  timeframe = 60,
  path_model,
  path_data,
  force_update = FALSE,
  objective_test = FALSE,
  num_nn_options = 12,
  fixed_nn_struct = c(100, 100),
  num_epoch = 100,
  num_bars_test = 600,
  num_bars_ahead = 34,
  num_cols_used = 16,
  min_perf = 0.3
)

Arguments

Details

Deep learning model structure is obtained from several random combinations of neurons within 3 hidden layers of the network. The most accurate model configuration will be automatically selected based either RMSE or Objective Test. In addition, the function will check if there is a need to update the model. To do that function will check results of the function aml_test_model.R.

Function is using the dataset prepared by the function aml_collect_data.R. Note that function will start to train the model as soon as there are more than 1000 rows in the dataset

Value

Function is writing a file object with the best Deep Learning Regression model

Author(s)

(C) 2020, 2021 Vladimir Zhbanko

Examples

library(dplyr)
library(readr)
library(h2o)
library(lazytrade)
library(lubridate)
library(magrittr)

path_model <- normalizePath(tempdir(),winslash = "/")
path_data <- normalizePath(tempdir(),winslash = "/")

ind = system.file("extdata", "AI_RSIADXUSDJPY60.csv",
                  package = "lazytrade") %>% read_csv(col_names = FALSE)

ind$X1 <- ymd_hms(ind$X1)

tick = system.file("extdata", "TickSize_AI_RSIADX.csv",
                  package = "lazytrade") %>% read_csv(col_names = FALSE)

write_csv(tick, file.path(path_data, "TickSize_AI_RSIADX.csv"), col_names = FALSE)

# data transformation using the custom function for one symbol
aml_collect_data(indicator_dataset = ind,
                 symbol = 'USDJPY',
                 timeframe = 60,
                 path_data = path_data)

# dataset will be written to the temp directory

# start h2o engine
h2o.init(nthreads = 2)


# performing Deep Learning Regression using 2 random neural network structures and objective test
aml_make_model(symbol = 'USDJPY',
               timeframe = 60,
               path_model = path_model,
               path_data = path_data,
               force_update=FALSE,
               objective_test = TRUE,
               num_nn_options = 6,
               num_epoch = 10,
               min_perf = 0,
               num_bars_test = 600,
               num_bars_ahead = 34,
               num_cols_used = 16)

# performing DL Regression using 2 random neural network structures
# with objective test, all columns
aml_make_model(symbol = 'USDJPY',
               timeframe = 60,
               path_model = path_model,
               path_data = path_data,
               force_update=FALSE,
               objective_test = TRUE,
               num_nn_options = 6,
               num_epoch = 10,
               min_perf = 0,
               num_bars_test = 600,
               num_bars_ahead = 34,
               num_cols_used = 0)


# performing Deep Learning Regression using the custom function
aml_make_model(symbol = 'USDJPY',
               timeframe = 60,
               path_model = path_model,
               path_data = path_data,
               force_update=FALSE,
               objective_test = FALSE,
               num_nn_options = 6,
               num_epoch = 10,
               min_perf = 0,
               num_bars_test = 600,
               num_bars_ahead = 34,
               num_cols_used = 16)

# performing Deep Learning Regression, fixed mode
aml_make_model(symbol = 'USDJPY',
               timeframe = 60,
               path_model = path_model,
               path_data = path_data,
               force_update=TRUE,
               num_nn_options = 0,
               fixed_nn_struct = c(100, 100),
               num_epoch = 10,
               min_perf = 0)

# stop h2o engine
h2o.shutdown(prompt = FALSE)

#set delay to insure h2o unit closes properly before the next test
Sys.sleep(5)

Function to score new data and predict change for each single currency pair

Description

Function is using the latest data from the financial assets indicator pattern and deep learning model. Prediction is a future price change for that asset

Usage

aml_score_data(symbol, timeframe, path_model, path_data, path_sbxm, path_sbxs)

Arguments

Details

Performs fresh data reading from the rds file

Value

Function is writing file into Decision Support System folder, mainly file with price change prediction in pips

Author(s)

(C) 2020 Vladimir Zhbanko

Examples

# test of function aml_make_model is duplicated here
library(dplyr)
library(readr)
library(lubridate)
library(h2o)
library(magrittr)
library(lazytrade)

path_model <- normalizePath(tempdir(),winslash = "/")
path_data <- normalizePath(tempdir(),winslash = "/")

ind = system.file("extdata", "AI_RSIADXUSDJPY60.csv",
                  package = "lazytrade") %>% read_csv(col_names = FALSE)

ind$X1 <- ymd_hms(ind$X1)

write_csv(ind, file.path(path_data, "AI_RSIADXUSDJPY60.csv"), col_names = FALSE)

# add tick data to the folder
tick = system.file("extdata", "TickSize_AI_RSIADX.csv",
                  package = "lazytrade") %>% read_csv(col_names = FALSE)

write_csv(tick, file.path(path_data, "TickSize_AI_RSIADX.csv"), col_names = FALSE)

# data transformation using the custom function for one symbol
aml_collect_data(indicator_dataset = ind,
                 symbol = 'USDJPY',
                 timeframe = 60,
                 path_data = path_data)
# start h2o engine (using all CPU's by default)

h2o.init(nthreads = 2)


# performing Deep Learning Regression using the custom function
aml_make_model(symbol = 'USDJPY',
               timeframe = 60,
               path_model = path_model,
               path_data = path_data,
               force_update=FALSE,
               num_nn_options = 2)


path_sbxm <- normalizePath(tempdir(),winslash = "/")
path_sbxs <- normalizePath(tempdir(),winslash = "/")


# score the latest data to generate predictions for one currency pair
aml_score_data(symbol = 'USDJPY',
               timeframe = 60,
               path_model = path_model,
               path_data = path_data,
               path_sbxm = path_sbxm,
               path_sbxs = path_sbxs)

# stop h2o engine
h2o.shutdown(prompt = FALSE)

#set delay to insure h2o unit closes properly before the next test
Sys.sleep(5)

Function to simulate multiple input structures

Description

Function is designed to evaluate several different inputs.

Usage

aml_simulation(
  timeframe = 60,
  path_sim_input,
  path_sim_result,
  par_simulate1 = 10,
  par_simulate2 = 16,
  demo_mode = FALSE
)

Arguments

Details

Function is using several other functions to perform sets of operations designed to test several inputs. Designed to validate model settings.

Update: New function structure to allow quicker simulation of parameters to find best performing input

Value

Function is writing file into Decision Support System folders

Author(s)

(C) 2021 Vladimir Zhbanko

Examples

library(dplyr)
library(magrittr)
library(readr)
library(h2o)
library(lazytrade)
library(lubridate)
library(stats)

path_input <- normalizePath(tempdir(),winslash = "/")
path_sim_input <- file.path(path_input, "path_sim_input")
dir.create(path_sim_input)
path_sim_result <- file.path(path_input, "path_sim_result")
dir.create(path_sim_result)

file.copy(from = system.file("extdata", "AI_RSIADXCADCHF60.csv", package = "lazytrade"),
          to = file.path(path_sim_input, "AI_RSIADXCADCHF60.csv"), overwrite = TRUE)
file.copy(from = system.file("extdata", "AI_RSIADXEURNZD60.csv", package = "lazytrade"),
          to = file.path(path_sim_input, "AI_RSIADXEURNZD60.csv"), overwrite = TRUE)

# start h2o engine
h2o.init(nthreads = 2)

# simulation of different epoch values
aml_simulation(timeframe = 60,
               path_sim_input = path_sim_input,
               path_sim_result = path_sim_result,
               par_simulate1 = 10,
               par_simulate2 = 10,
               demo_mode = FALSE)

Sys.sleep(5)
# stop h2o engine
h2o.shutdown(prompt = FALSE)

#set delay to insure h2o unit closes properly before the next test
Sys.sleep(5)

Function to test the model and conditionally decide to update existing model for a single currency pair

Description

Function is designed to test the trading decision generated by the Deep learning regression model. It is doing so by simulating trading strategy outcome. The outcome of this function will be also used to define best trigger to join into the trading opportunity

Usage

aml_test_model(
  symbol,
  num_bars,
  timeframe,
  path_model,
  path_data,
  path_sbxm = path_sbxm,
  path_sbxs = path_sbxs
)

Arguments

Details

Function is reading price data and corresponding indicator. Starting from the trained model function will test the trading strategy using simplified trading approach. Trading approach will entail using the last available indicator data, predict the price change for every row, trade will be simulating by holding the asset for 3, 5, 10 and 34 hours. Several trigger points will be evaluated selecting the most optimal trading trigger. Function is writing most optimal decision into *.csv file Such file will be used by the function aml_make_model.R to decide whether model must be updated...

Value

Function is writing file into Decision Support System folders

Author(s)

(C) 2020, 2021 Vladimir Zhbanko

Examples

library(dplyr)
library(magrittr)
library(readr)
library(h2o)
library(lazytrade)
library(lubridate)

path_model <- normalizePath(tempdir(),winslash = "/")
path_data <- normalizePath(tempdir(),winslash = "/")

ind = system.file("extdata", "AI_RSIADXUSDJPY60.csv",
                  package = "lazytrade") %>% read_csv(col_names = FALSE)

ind$X1 <- ymd_hms(ind$X1)

tick = system.file("extdata", "TickSize_AI_RSIADX.csv",
                  package = "lazytrade") %>% read_csv(col_names = FALSE)

write_csv(ind, file.path(path_data, "AI_RSIADXUSDJPY60.csv"), col_names = FALSE)

write_csv(tick, file.path(path_data, "TickSize_AI_RSIADX.csv"), col_names = FALSE)

# data transformation using the custom function for one symbol
aml_collect_data(indicator_dataset = ind,
                 symbol = 'USDJPY',
                 timeframe = 60,
                 path_data = path_data)


# start h2o engine
h2o.init(nthreads = 2)


# performing Deep Learning Regression using the custom function
aml_make_model(symbol = 'USDJPY',
               timeframe = 60,
               path_model = path_model,
               path_data = path_data,
               force_update=FALSE,
               num_nn_options = 3)

path_sbxm <- normalizePath(tempdir(),winslash = "/")
path_sbxs <- normalizePath(tempdir(),winslash = "/")

# score the latest data to generate predictions for one currency pair
aml_score_data(symbol = 'USDJPY',
               timeframe = 60,
               path_model = path_model,
               path_data = path_data,
               path_sbxm = path_sbxm,
               path_sbxs = path_sbxs)

# test the results of predictions
aml_test_model(symbol = 'USDJPY',
               num_bars = 600,
               timeframe = 60,
               path_model = path_model,
               path_data = path_data,
               path_sbxm = path_sbxm,
               path_sbxs = path_sbxs)

# stop h2o engine
h2o.shutdown(prompt = FALSE)

#set delay to insure h2o unit closes properly before the next test
Sys.sleep(5)

Function check_if_optimize.

Description

Purpose of this function is to verify trading system functionality by analysing profit factor on the last trades. Whenever trading robot has profit factor value below certain limit function will write a file log indicating which trading systems need to be maintained.

Learn by example how to manipulate data

Usage

check_if_optimize(
  x,
  system_list,
  path_data,
  num_trades_to_consider = 3,
  profit_factor_limit = 0.7,
  write_mode = FALSE
)

Arguments

Details

Whenever there will be not enough trades then empty file will be written to the destination

Value

function returns a dataframe with systems that should be optimized

Author(s)

(C) 2019,2021 Vladimir Zhbanko

Examples

library(lazytrade)
library(magrittr)
library(dplyr)
library(readr)
library(lubridate)

path_data <- normalizePath(tempdir(),winslash = "/")

file.copy(from = system.file("extdata", "Setup.csv", package = "lazytrade"),
          to = file.path(path_data, "Setup.csv"), overwrite = TRUE)

system_list <- read_csv(file.path(path_data, "Setup.csv"))

data(profit_factorDF)


# without writing to the file
check_if_optimize(x = profit_factorDF,
                  system_list = system_list,
                  path_data,
                  num_trades_to_consider = 3,
                  profit_factor_limit = 0.8,
                  write_mode = TRUE)

Create labelled data

Description

FUNCTION create_labelled_data. PURPOSE: function gets price data of every currency in each column. It is splitting this data by periods and transposes the data. Additionally function is capable to label the data based on the simple logic. Each row will be assigned into 2 categories based on the difference between beginning and end of the row elements Finally all data will be stacked on top and joined into the table

Learn by example how to manipulate data

Usage

create_labelled_data(x, n = 50, type = "regression")

Arguments

Details

see more info in the udemy course self-learning-trading-robot

Value

function returns transposed data. One column called 'LABEL' indicate achieved value of the label. Transposed values from every column are stacked one to each other

Examples

library(dplyr)
library(magrittr)
library(readr)
library(lazytrade)

# usind a sample data
data(price_dataset)

# price change as a label
create_labelled_data(x = price_dataset, n = 75, type = "regression")

# factors 'BU'/'BE' as a label
create_labelled_data(x = price_dataset, n = 75, type = "classification")

Create Transposed Data

Description

PURPOSE: function gets indicator data in each column. Goal is to splitting this data into periods and transpose the data.

Usage

create_transposed_data(x, n = 50)

Arguments

Details

each column contains records of the indicator value of the assets every column will be split into chunks of n observations and transposed into rows this repeated for all the columns coming up with a matrix. Function works in combination with a function create_labelled_data

Value

function returns transposed data. Transposed values from every column are stacked one to each other

Examples

library(dplyr)
library(magrittr)
library(lazytrade)

# usind a sample data
data(indicator_dataset)

create_transposed_data(indicator_dataset, n = 75)

Table with Trade results samples

Description

Table with Trade results samples

Usage

data_trades

Format

A dataframe with several columns

MagicNumber

Unique identifiers of the Trading Robots

TicketNumber

Ticket Number of closed position

OrderStartTime

Date and Time when order started

OrderCloseTime

Date and Time when order closed

Profit

Monetary result of the trade

Symbol

Symbol of the Asset e.g. EURUSD

OrderType

Order Type 0 - buy, 1 - sell

Function that decrypt encrypted content

Description

Cryptography facility

Usage

decrypt_mykeys(path_encrypted_content, path_private_key)

Arguments

Details

It is possible to generate private/public key pair using R-Studio Project Options Menu. Alternatively possible to use 'openssl' R package

Value

• a string with decrypted key

Examples

library(dplyr)
library(magrittr)
library(openssl)
library(readr)

path_ssh <- normalizePath(tempdir(),winslash = "/")
rsa_keygen() %>% write_pem(path = file.path(path_ssh, 'id_api'))
# extract and write your public key
read_key(file = file.path(path_ssh, 'id_api'), password = "") %>%
`[[`("pubkey") %>% write_pem(path = file.path(path_ssh, 'id_api.pub'))

path_private_key <- file.path(path_ssh, "id_api")
path_public_key <- file.path(path_ssh, "id_api.pub")

#encrypting string 'my_key'...
encrypt_api_key(api_key = 'my_key', enc_name = 'api_key.enc.rds',path_ssh = path_ssh)

#encrypted content
out <- read_rds(file.path(path_ssh, "api_key.enc.rds"))

# Consumer API keys
ConsumerAPIkeys <- decrypt_mykeys(path_encrypted_content = file.path(path_ssh,
                                  'api_key.enc.rds'),
                                  path_private_key = path_private_key)

Create log difference distribution

Description

Calculate log distribution and calculate difference within rows

Usage

dlog(x)

Arguments

Value

• dataframe

Examples

library(magrittr)
library(lazytrade)
m <- seq(1:1000) %>% as.matrix(10) %>% dlog()

Encrypt api keys

Description

Provide easy interface to encrypt the api key. In order to use function simply provide a string with an API key. In addition provide the path to the .ssh folder and names of the private and public keys

Usage

encrypt_api_key(
  api_key,
  enc_name = "api_key.enc.rds",
  path_ssh = "path_ssh",
  file_rsa = "id_api",
  file_rsa_pub = "id_api.pub"
)

Arguments

Details

Make sure to clean the history of the R session

Value

Writes a file with encrypted key

References

for more info on how to use RSA cryptography in R check my course on Udemy

Examples

library(openssl)
library(magrittr)
library(readr)
path_ssh <- normalizePath(tempdir(),winslash = "/")
rsa_keygen() %>% write_pem(path = file.path(path_ssh, 'id_api'))
# extract and write your public key
read_key(file = file.path(path_ssh, 'id_api'), password = "") %>%
`[[`("pubkey") %>% write_pem(path = file.path(path_ssh, 'id_api.pub'))


path_private_key <- file.path(path_ssh, "id_api")
path_public_key <- file.path(path_ssh, "id_api.pub")

#encrypting string 'my_key'...
encrypt_api_key(api_key = 'my_key', enc_name = 'api_key.enc.rds',path_ssh = path_ssh)

out <- read_rds(file.path(path_ssh, "api_key.enc.rds"))

# decrypting the password using public data list and private key
api_key <- decrypt_envelope(out$data,
                            out$iv,
                            out$session,
                            path_private_key, password = "") %>%
           unserialize()

# outcome of the encryption will be a string 'my_key'

Function used to evaluate market type situation by reading the file with Macroeconomic Events and writing a trigger to the trading robot

Description

Function is reading the content of the file 01_MacroeconomicEvent.csv. Content of the file can be either 1 or 0. 1 - when Macro Economic event is present, 0 - when it's not. Function will also read magic number of the trading robots. This is indicated in the file 'Setup.csv'. Final outcome of the function is the series of files written to the destination directories. These files will either enable or disable opening of new positions in the trading robots

Usage

evaluate_macroeconomic_event(
  setup_file_path,
  setup_file_name = "Setup.csv",
  macro_event_path,
  macro_file_name = "01_MacroeconomicEvent.csv",
  path_T1,
  path_T3
)

Arguments

Details

This function is used exclusively with Market Type recognition system.

Final evaluation will consist in writing a dedicated file with a simple information:

When Macro economic even is not present:

"Magic","IsEnabled" 8139125,1

or, when Macro economic event is present:

"Magic","IsEnabled" 8139125,0

Value

Function will write files indicating to enable or disable trading systems to open new orders

Examples

# evaluate data on macroeconomic event (required to start trading)
library(dplyr)
library(readr)
library(lazytrade)

dir <- normalizePath(tempdir(),winslash = "/")

evaluate_macroeconomic_event(setup_file_path = system.file('extdata', package = "lazytrade"),
                             setup_file_name = "Setup.csv",
                             macro_event_path = system.file('extdata', package = "lazytrade"),
                             macro_file_name = "01_MacroeconomicEvent.csv",
                             path_T1 = dir, path_T3 = dir)

Function that returns the profit factors of the systems in a form of a DataFrame

Description

Calculation of profit factor using dplyr verbs

Usage

get_profit_factorDF(x, num_orders)

Arguments

Value

• Function returns dataframe with column PrFact with calculated profit factor value for each trading robot

Examples

library(lazytrade)
library(dplyr)
library(magrittr)
data(profit_factorDF)
get_profit_factorDF(x = profit_factorDF,
                    num_orders = 10)

Import Data file with Trade Logs to R.

Description

Function is capable to import file with executed trades log. Files do not have column headers hence function will take care to name columns as well as to perform relevant cleansing

Usage

import_data(path_sbxm, trade_log_file)

Arguments

Value

Function will return the dataframe with trade data and automatically set proper column types

Author(s)

(C) 2019, 2020 Vladimir Zhbanko

Examples

library(lazytrade)
library(dplyr)
library(readr)
library(lubridate)

path_sbxm <- normalizePath(tempdir(),winslash = "/")

file.copy(from = system.file("extdata", "OrdersResultsT1.csv", package = "lazytrade"),
          to = file.path(path_sbxm, "OrdersResultsT1.csv"), overwrite = TRUE)

DFT1 <- import_data(path_sbxm = path_sbxm,
                    trade_log_file = "OrdersResultsT1.csv")

Table with indicator dataset

Description

Table with indicator dataset

Usage

indicator_dataset

Format

A dataframe with several columns

X1

Date and time of the indicator sample

X2-X29

Values of the assets

Table with indicator only used to train model, 128 col 1646 rows

Description

Table with indicator only used to train model, 128 col 1646 rows

Usage

macd_100

Format

A dataframe with several columns

EURUSD

Values of the macd indicator

GBPUSD

Values of the macd indicator

AUDUSD

Values of the macd indicator

NZDUSD

Values of the macd indicator

USDCAD

Values of the macd indicator

USDCHF

Values of the macd indicator

USDJPY

Values of the macd indicator

EURGBP

Values of the macd indicator

EURJPY

Values of the macd indicator

EURCHF

Values of the macd indicator

EURNZD

Values of the macd indicator

EURCAD

Values of the macd indicator

EURAUD

Values of the macd indicator

GBPAUD

Values of the macd indicator

GBPCAD

Values of the macd indicator

GBPCHF

Values of the macd indicator

GBPJPY

Values of the macd indicator

GBPNZD

Values of the macd indicator

AUDCAD

Values of the macd indicator

AUDCHF

Values of the macd indicator

AUDJPY

Values of the macd indicator

AUDNZD

Values of the macd indicator

CADJPY

Values of the macd indicator

CHFJPY

Values of the macd indicator

NZDJPY

Values of the macd indicator

NZDCAD

Values of the macd indicator

NZDCHF

Values of the macd indicator

CADCHF

Values of the macd indicator

Table with indicator and market type category used to train model

Description

Table with indicator and market type category used to train model

Usage

macd_ML60M

Format

A dataframe with several columns

X1-X64

Values of the macd indicator

M_T

Category of Market Type

Table with one column indicator dataset

Description

Table with one column indicator dataset

Usage

macd_df

Format

A dataframe with one column

CADCHF

Indicator values of the asset

Function to score data and predict current market type using pre-trained classification model

Description

PURPOSE: Function that uses Deep Learning model and Time Series Column of the dataframe to find out specific market type of the financial asset it will also discard bad result outputting -1 if it is the case

Usage

mt_evaluate(x, path_model, num_cols, timeframe)

Arguments

Details

it is mandatory to switch on the virtual h2o machine with h2o.init() also to shut it down with h2o.shutdown(prompt = F)

Value

dataframe with predicted value of the market type

Examples

library(h2o)
library(magrittr)
library(dplyr)
library(readr)
library(lazytrade)

path_model <- normalizePath(tempdir(),winslash = "/")
path_data <- normalizePath(tempdir(),winslash = "/")

data(macd_ML60M)

# start h2o engine (using all CPU's by default)
h2o.init(nthreads = 2)

# performing Deep Learning Regression using the custom function
# this function stores model to the temp location
mt_make_model(indicator_dataset = macd_ML60M,
              num_bars = 64,
              timeframe = 60,
              path_model = path_model,
              path_data = path_data,
              activate_balance = TRUE,
              num_nn_options = 3)


# Use sample data
data(macd_100)

# use one column for testing
x <- macd_100[ ,2]


mt_evaluate(x = x,
            path_model = path_model,
            num_cols = 64,
            timeframe = 60)

h2o.shutdown(prompt = FALSE)

#set delay to insure h2o unit closes properly before the next test
Sys.sleep(5)

Import Market Type related Data to R from the Sandbox

Description

Function imports file from the MetaTrader sandbox. Function performs necessary cleansing of the data column types

Usage

mt_import_data(path_sbxm, system_number)

Arguments

Value

function returns the data frame with 5 columns including market type code

Author(s)

(C) 2020, 2021 Vladimir Zhbanko

Examples

library(dplyr)
library(readr)
library(lazytrade)

path_sbxm <- normalizePath(tempdir(),winslash = "/")

file.copy(from = system.file("extdata", "MarketTypeLog9139106.csv", package = "lazytrade"),
          to = file.path(path_sbxm, "MarketTypeLog9139106.csv"), overwrite = TRUE)

DF1 <- mt_import_data(path_sbxm = path_sbxm,
                      system_number = 9139106)

Function to train Deep Learning Classification model for Market Type recognition

Description

Function is training h2o deep learning model to match classified patterns of the financial indicator. Main idea is to be able to detect Market Type by solely relying on the current indicator pattern. This is in the attempt to evaluate current market type for trading purposes.

Selected Market Periods could be manually classified according to the theory from Van K. Tharp:

1. Bull normal, BUN

2. Bull volatile, BUV

3. Bear normal, BEN

4. Bear volatile, BEV

5. Sideways quiet, RAN

6. Sideways volatile, RAV

For automatic classification, could only be used: BUN, BEN, RAN market types

Usage

mt_make_model(
  indicator_dataset,
  num_bars = 64,
  timeframe = 60,
  path_model,
  path_data,
  activate_balance = TRUE,
  num_nn_options = 24,
  fixed_nn_struct = c(100, 100),
  num_epoch = 100,
  is_cluster = FALSE
)

Arguments

Details

Function is using labeled dataset and tries several different random neural network structures. Once the best neural network is found then the better model is selected and stored. Dataset can be either manually labelled or generated using function mt_stat_transf.R. In the latter case parameter is_cluster shall be set to TRUE.

Value

Function is writing file object with the model

Author(s)

(C) 2020, 2021 Vladimir Zhbanko

Examples

library(dplyr)
library(magrittr)
library(readr)
library(h2o)
library(lazytrade)
library(stats)

path_model <- normalizePath(tempdir(),winslash = "/")
path_data <- normalizePath(tempdir(),winslash = "/")

data(macd_ML60M)

Sys.sleep(5)

# start h2o engine
h2o.init(nthreads = 2)


# performing Deep Learning Classification using manually labelled data
mt_make_model(indicator_dataset = macd_ML60M,
              num_bars = 64,
              timeframe = 60,
              path_model = path_model,
              path_data = path_data,
              activate_balance = TRUE,
              num_nn_options = 3,
              num_epoch = 10)

data(price_dataset_big)
data <- head(price_dataset_big, 5000) #reduce computational time

ai_class <- mt_stat_transf(indicator_dataset = data,
                      num_bars = 64,
                      timeframe = 60,
                      path_data = path_data,
                      mt_classes = c('BUN', 'BEN', 'RAN'))

# performing Deep Learning Classification using the custom function auto clustered data
mt_make_model(indicator_dataset = ai_class,
              num_bars = 64,
              timeframe = 60,
              path_model = path_model,
              path_data = path_data,
              activate_balance = TRUE,
              num_nn_options = 6,
              num_epoch = 10,
              is_cluster = TRUE)

# performing Deep Learning Classification using the custom function auto clustered data
# and fixed nn structure
mt_make_model(indicator_dataset = ai_class,
              num_bars = 64,
              timeframe = 60,
              path_model = path_model,
              path_data = path_data,
              activate_balance = TRUE,
              num_nn_options = 0,
              fixed_nn_struct = c(10, 10),
              num_epoch = 10,
              is_cluster = TRUE)
              
# stop h2o engine
h2o.shutdown(prompt = FALSE)

#set delay to insure h2o unit closes properly before the next test
Sys.sleep(5)

Function to prepare and score data, finally predict current market type using pre-trained classification model

Description

PURPOSE: Function that uses Deep Learning model and Time Series Column of the dataframe to find out specific market type of the financial asset it will also discard bad result outputting -1 if it is the case

Usage

mt_stat_evaluate(x, path_model, num_bars, timeframe)

Arguments

Value

dataframe with predicted value of the market type

Author(s)

(C) 2021 Vladimir Zhbanko

Examples

library(h2o)
library(magrittr)
library(dplyr)
library(readr)
library(lazytrade)
library(stats)

path_model <- normalizePath(tempdir(),winslash = "/")
path_data <- normalizePath(tempdir(),winslash = "/")


# start h2o engine (using all CPU's by default)
h2o.init(nthreads = 2)

data(price_dataset_big)
data <- head(price_dataset_big, 500) #reduce computational time

ai_class <- mt_stat_transf(indicator_dataset = data,
                      num_bars = 64,
                      timeframe = 60,
                      path_data = path_data,
                      mt_classes = c('BUN', 'BEN', 'RAN'))

# performing Deep Learning Classification using the custom function auto clustered data
mt_make_model(indicator_dataset = ai_class,
              num_bars = 64,
              timeframe = 60,
              path_model = path_model,
              path_data = path_data,
              activate_balance = TRUE,
              num_nn_options = 3,
              num_epoch = 10,
              is_cluster = TRUE)


# Use sample data
data(price_dataset)

# use one column for testing
x <- price_dataset[ ,2]


mt_stat_evaluate(x = x,
                 path_model = path_model,
                 num_bars = 64,
                 timeframe = 60)

h2o.shutdown(prompt = FALSE)

#set delay to insure h2o unit closes properly before the next test
Sys.sleep(5)

Perform Statistical transformation and clustering of Market Types on the price data

Description

Function features methods of statistical data transformation and clustering of the price data. Multiple statistical properties are calculated for a defined time interval. Once combined, unsupervised learning (clustering) is performed to assign several classes, see function mt_make_model. Function allows to fully automatize financial periods classification. It is possible to choose two clustering methods either kmeans or hierarchical clustering.

Usage

mt_stat_transf(
  indicator_dataset,
  num_bars = 64,
  timeframe = 60,
  path_data,
  mt_classes,
  clust_method = "kmeans",
  clust_opt = "complete",
  rule_opt = TRUE
)

Arguments

Details

User can define several market type classes names however function will randomly assign Market Period labels based on Unsupervised Learning. This is inconvenient however that should be compensated by automated way of doing such data classification

Value

Dataframe with statistically transformed and classified dataset for classification modeling

Author(s)

(C) 2021 Vladimir Zhbanko

Examples

library(dplyr)
library(stats)
library(magrittr)
library(readr)
library(lazytrade)

path_data <- normalizePath(tempdir(),winslash = "/")

data(price_dataset_big)

#option
#mt_classes = c('BUN', 'BEN', 'RAN','BUV', 'BEV', 'RAV')
#mt_classes = c('BUN', 'BEN', 'RAN')
#clust_method = 'hclust'
#clust_opt = 'ward'

#build dataset for Market Type detection without rule based check
ai_class_rand <- mt_stat_transf(indicator_dataset = price_dataset_big,
                                num_bars = 64,
                                timeframe = 60,
                                path_data = path_data,
                                mt_classes = c('BUN', 'BEN'),
                                clust_method = 'kmeans',
                                clust_opt = 'complete',
                                rule_opt = FALSE)

#use rule base check
ai_class_rule <- mt_stat_transf(indicator_dataset = price_dataset_big,
                                num_bars = 64,
                                timeframe = 60,
                                path_data = path_data,
                                mt_classes = c('BUN', 'BEN'),
                                clust_method = 'kmeans',
                                clust_opt = 'complete',
                                rule_opt = TRUE)

Function to aggregate trading results from multiple folders and files

Description

Read multiple '.csv' files stored in different folders Store results to the intermediate dataframe.

Usage

opt_aggregate_results(path_data)

Arguments

Details

user must provide the path to the files in the folders all files in subfolders are read and aggregated into one data object. Data object is sorted in descending order by order close time

Value

Dataframe with trading results

Examples

 library(lazytrade)
 library(readr)
 library(dplyr)
 library(magrittr)
 library(lubridate)

 dir <- normalizePath(tempdir(),winslash = "/")

 file.copy(from = system.file("extdata/RES", package = "lazytrade"),
           to = dir, recursive = TRUE)

 DF_RES <- opt_aggregate_results(path_data = file.path(dir, "RES"))

Function to create summary graphs of the trading results

Description

Create graphs and store them into pdf file

Usage

opt_create_graphs(x, outp_path, graph_type = "pdf")

Arguments

Details

bar graph and time series optionally written to the pdf file. File is named with a date of analysis to the location specified by the user

Value

graphic output

Examples

 library(lazytrade)
 library(readr)
 library(dplyr)
 library(magrittr)
 library(lubridate)
 library(ggplot2)
 data(DFR)
 dir <- normalizePath(tempdir(),winslash = "/")
 # create pdf file with two graphs
 opt_create_graphs(x = DFR, outp_path = dir)

 # only show time series plot
 opt_create_graphs(x = DFR, graph_type = 'ts')

Table with Market Types and sample of actual policy for those states

Description

Table with Market Types and sample of actual policy for those states

Usage

policy_tr_systDF

Format

A dataframe with 2 columns:

MarketType

Current Market Type status

Policy

Policy choice

Table with price dataset

Description

Table with price dataset

Usage

price_dataset

Format

A dataframe with several columns

X1

Date and time of the price sample

X2-X29

Values of the assets

Table with price dataset, 30000 rows

Description

Table with price dataset, 30000 rows

Usage

price_dataset_big

Format

A dataframe with several columns

X1

Date and time of the price sample

X2-X29

Values of the assets

Table with Trade results samples

Description

Table with Trade results samples

Usage

profit_factorDF

Format

A dataframe with several columns

MagicNumber

Unique identifiers of the Trading Robots

TicketNumber

Ticket Number of closed position

OrderStartTime

Date and Time when order started

OrderCloseTime

Date and Time when order closed

Profit

Monetary result of the trade

Symbol

Symbol of the Asset e.g. EURUSD

OrderType

Order Type 0 - buy, 1 - sell

Table with Trade results samples

Description

Table with Trade results samples

Usage

profit_factor_data

Format

A dataframe with several columns

X1

Unique identifiers of the Trading Robots

X2

Ticket Number of closed position

X3

Date and Time when order started

X4

Date and Time when order closed

X5

Monetary result of the trade

X6

Symbol of the Asset e.g. EURUSD

X7

Order Type 0 - buy, 1 - sell

Table with predicted price change

Description

Table with predicted price change

Usage

result_R

Format

A dataframe with one column

predict

predicted future price change

Table with aggregated trade results

Description

Table with aggregated trade results

Usage

result_R1

Format

A dataframe with one column

predict

predicted price change

Table with one column as result from the model prediction

Description

Table with one column as result from the model prediction

Usage

result_prev

Format

A dataframe with one column

predict

Predicted values from the model

Function performs Reinforcement Learning using the past data to generate model policy

Description

This function will perform Reinforcement Learning using Trading Data. It will suggest whether or not it is better to keep using trading systems or not. Function is just using results of the past performance to generate the recommendation (not a holy grail).

Usage

rl_generate_policy(x, states, actions, control)

Arguments

Details

Initial policy is generated using a dummy zero values. This way function starts working directly from the first observation. However policy 'ON' value will only be generated once the Q value is greater than zero

Value

Function returns data frame with reinforcement learning model policy

Author(s)

(C) 2019,2021 Vladimir Zhbanko

Examples

  library(dplyr)
  library(ReinforcementLearning)
  library(magrittr)
  library(lazytrade)

  data(data_trades)
  states <- c("tradewin", "tradeloss")
  actions <- c("ON", "OFF")
  control <- list(alpha = 0.7, gamma = 0.3, epsilon = 0.1)
  rl_generate_policy(x = data_trades,
                     states, actions, control)

Function performs RL and generates model policy for each Market Type

Description

This function will perform Reinforcement Learning using Trading Data. It will suggest whether or not it is better to keep using trading systems or not. Function is just using results of the past performance to generate the recommendation (not a holy grail).

Usage

rl_generate_policy_mt(x, states, actions, control)

Arguments

Details

Initial policy is generated using a dummy zero values. This way function starts working directly from the first observation. However policy 'ON' value will only be generated once the Q value is greater than zero

Value

Function returns data frame with reinforcement learning model policy

Examples

library(dplyr)
library(magrittr)
library(ReinforcementLearning)
library(lazytrade)
data(trading_systemDF)
states <- c("BUN", "BUV", "BEN", "BEV", "RAN", "RAV")
actions <- c("ON", "OFF")
control <- list(alpha = 0.7, gamma = 0.3, epsilon = 0.1)
rl_generate_policy_mt(x = trading_systemDF,
                      states = states,
                      actions = actions,
                      control = control)

Function to retrieve and help to log Q values during RL progress.

Description

Function will record Q values during the model update. These values will be used by another function Function was developed to help to estimate best control parameters during optimisation process

Usage

rl_log_progress(x, states, actions, control)

Arguments

Value

dataframe with log of RL model reward sequences during model update

Examples

# retrieve RL model Q values progress
library(ReinforcementLearning)
library(dplyr)
library(magrittr)
library(lazytrade)
data(data_trades)
x <- data_trades
states <- c("tradewin", "tradeloss")
actions <- c("ON", "OFF")
control <- list(alpha = 0.7, gamma = 0.3, epsilon = 0.1)

rl_log_progress(x = x,states = states, actions = actions, control = control)

Function to retrieve and help to log Q values during RL progress. This function is dedicated to the situations when Market Types are used as a 'states' for the Environment.

Description

Function will record Q values during the model update. These values will be used by another function Function was developed to help to estimate best control parameters during optimisation process

Usage

rl_log_progress_mt(x, states, actions, control)

Arguments

Value

dataframe with log of RL model reward sequences during model update

Author(s)

(C) 2020, 2021 Vladimir Zhbanko

Examples

# retrieve RL model Q values progress
library(ReinforcementLearning)
library(dplyr)
library(magrittr)
library(lazytrade)
data(trading_systemDF)
x <- trading_systemDF
states <- c("BUN", "BUV", "BEN", "BEV", "RAN", "RAV")
actions <- c("ON", "OFF") # 'ON' and 'OFF' are referring to decision to trade with Slave system
control <- list(alpha = 0.7, gamma = 0.3, epsilon = 0.1)

rl_log_progress_mt(x = x,states = states, actions = actions, control = control)

Record Reinforcement Learning Policy.

Description

Function will write a policy 'decision' to the csv file specific for each Expert Advisor

Usage

rl_record_policy(
  x,
  last_result,
  trading_system,
  path_terminal,
  fileName = "SystemControl"
)

Arguments

Details

It is imperative that terminal path contains exact word Terminal3

Value

nothing is returned but function will write csv file to the supplied directory

Author(s)

(C) 2019,2021 Vladimir Zhbanko

Examples

library(stringr)
library(magrittr)
library(dplyr)
data(TradeStatePolicy)

dir <- normalizePath(tempdir(),winslash = "/")

rl_record_policy(x = TradeStatePolicy,
              last_result = "tradewin",
              trading_system = 8118101,
              path_terminal = dir,
              fileName = "SystemControlRL")

Record Reinforcement Learning Policy for Market Types

Description

Function will write a policy 'decision' to the csv file specific for each Expert Advisor

Usage

rl_record_policy_mt(
  x,
  trading_system,
  path_terminal,
  fileName = "SystemControlMT"
)

Arguments

Details

It is imperative that terminal path contains exact word Terminal3

Value

nothing is returned but function will write csv file to the supplied directory

Examples

library(stringr)
library(lazytrade)
data(policy_tr_systDF)

dir <- normalizePath(tempdir(),winslash = "/")

rl_record_policy_mt(x = policy_tr_systDF,
                 trading_system = 8118101,
                 path_terminal = dir,
                 fileName = "SystemControlMT")

Function to find and write the best control parameters.

Description

This function is supposed to run on a weekly basis. Purpose of this function is to perform RL and trading simulation and find out the best possible control parameters for the RL function.

Usage

rl_write_control_parameters(
  x,
  path_control_files,
  num_trades_to_consider = 100
)

Arguments

Details

Function is used by the R script Adapt_RL_control.R

Value

Function writes best control parameters to be used by the Reinforcement Learning Function

Author(s)

(C) 2019 Vladimir Zhbanko

Examples

dir <- normalizePath(tempdir(),winslash = "/")
#test lasts 15 sec:
library(dplyr)
library(readr)
library(ReinforcementLearning)
library(magrittr)
library(lazytrade)
data(data_trades)
x <- data_trades
rl_write_control_parameters(x = data_trades,
                            path_control_files = dir,
                            num_trades_to_consider = 20)

Function to find and write the best control parameters.

Description

This function is supposed to run on a weekly basis. Purpose of this function is to perform RL and trading simulation and find out the best possible control parameters for the RL function.

Usage

rl_write_control_parameters_mt(
  x,
  path_control_files,
  num_trades_to_consider = 100
)

Arguments

Details

Function is used by the R script Adapt_RL_MT_control.R

Value

Function writes best control parameters to be used by the Reinforcement Learning Function

Author(s)

(C) 2019, 2021 Vladimir Zhbanko

Examples

# test lasts 15 sec:
dir <- normalizePath(tempdir(),winslash = "/")

library(dplyr)
library(readr)
library(ReinforcementLearning)
library(magrittr)
library(lazytrade)
data(trading_systemDF)

# use optimal control parameters found by auxiliary function
rl_write_control_parameters_mt(x = trading_systemDF,
                               path_control_files = dir,
                               num_trades_to_consider = 100)

Table with several columns containing indicator values and Label values

Description

Table with several columns containing indicator values and Label values

Usage

test_data_pattern

Format

A dataframe with several columns

LABEL

Asset values as were recorded in the future

V1-V49

Transposed values of the indicator

Convert time series data to matrix with defined number of columns

Description

Transforms Time Series Column of the dataframe to the matrix with specified number of columns. Number of rows will be automatically found. Eventually not complete last row will be discarded.

Usage

to_m(x, n_cols)

Arguments

Value

• matrix with specified amount of rows

Examples

library(magrittr)
library(lazytrade)
macd_m <- seq(1:1000) %>% as.data.frame() %>% to_m(64)

Table with trade data and joined market type info

Description

Table with trade data and joined market type info

Usage

trading_systemDF

Format

A dataframe with several columns

"MagicNumber.x

Unique identifiers of the Trading Robots from Trade Log

TicketNumber

Ticket Number of closed position

OrderStartTime

Date and Time when order started

OrderCloseTime

Date and Time when order closed

Profit

Monetary result of the trade

Symbol

Symbol of the Asset e.g. EURUSD

OrderType

Order Type 0 - buy, 1 - sell

"MagicNumber.y

Unique identifiers of the Trading Robots from Ticket Opening Log

"MarketType

Logged Market Type of the asset at the moment of Ticket Opening

R function to find file with specific code within it's content

Description

Utility function to find the file path which contains specific code

Usage

util_find_file_with_code(
  files,
  code_to_find,
  option_replace = FALSE,
  v_settings,
  v_values
)

Arguments

Details

Function is reading content of file and checks each file for a specific code Function will require a vector with a valid file paths Function will output the path to that file which contains a give code If no desired code is found, function will return NULL Optionally, if enabled, function will replace their values Note: supply same kind of parameters (either boolean or integers)

Value

string with a file path containing a code

Author(s)

(C) 2024 Vladimir Zhbanko

Examples

library(readr)

dir <- normalizePath(tempdir(),winslash = "/")

file.copy(from = system.file("extdata/default", package = "lazytrade"),
          to = dir, recursive = TRUE)

files_chr <- list.files(path = file.path(dir, "default"), all.files = TRUE,
pattern = ".chr", full.names = TRUE)
code_to_find <- 9142301

# find the path to the file with numeric code
file_i_need <- util_find_file_with_code(files = files_chr, 
                                   code_to_find = 9142301)

# find the path to the file with a character code
file_i_need <- util_find_file_with_code(files = files_chr,
                                   code_to_find = 'BITCOIN')
                                   
# find the path to the file with a character code that is not exists in those files
file_i_need <- util_find_file_with_code(files = files_chr,
                                   code_to_find = 'Banana')
                                   
# define a vector with settings to search
v_par <- c("StartHour")
v_val <- c(15)
                                    
# Replace integer values
file_i_need <- util_find_file_with_code(files = files_chr,
                                   code_to_find = 9142301,
                                   option_replace = TRUE,
                                   v_settings = v_par,
                                   v_values = v_val)
                                   
# define a vector with settings to search
v_par <- "UseMAFilter"
v_val <- FALSE
                                     
# Replace boolean values
file_i_need <- util_find_file_with_code(files = files_chr,
                                   code_to_find = 9142301,
                                   option_replace = TRUE,
                                   v_settings = v_par,
                                   v_values = v_val)
                                     
# Replace boolean values in specified file
file.copy(from = system.file("extdata/Falcon_D.set", package = "lazytrade"),
          to = dir, recursive = TRUE)
file_i_need <- util_find_file_with_code(files = file.path(dir, "Falcon_D.set"),
                                   code_to_find = 9142301,
                                   option_replace = TRUE,
                                   v_settings = v_par,
                                   v_values = v_val)
           

R function to find PID of active applications

Description

Utility function to find PID of the working terminal.exe application Function is created to generate a system call to programmatically close any given application

Usage

util_find_pid(tasks_running = t_running, pid_pattern = "terminal.exe")

Arguments

Details

Function is executing a system command to get all processes running on the OS Retrieved data is cleaned and organized to filter on required process Function can also be used to track specific applications defined by the user

Value

string with system kill command to close selected application

Author(s)

(C) 2021 Vladimir Zhbanko

Examples

library(magrittr)
library(tibble)
library(stringr)
library(dplyr)

#library(readr)


dir <- normalizePath(tempdir(),winslash = "/")


#tasks_running <- system("tasklist", intern = TRUE)
#writeLines(tasks_running, con = file.path(dir,'tasks_running.txt'))
#t_running <- readLines(con = file.path(dir,'tasks_running.txt'))

tasks_list = system.file("extdata", "tasks_running.txt",
                  package = "lazytrade")

t_running <- readLines(con = tasks_list)

#generate task kill command for this application
util_find_pid(tasks_running = t_running,
              pid_pattern = 'terminal.exe')

util_find_pid(tasks_running = t_running,
              pid_pattern = 'chrome.exe')

R function to generate random passwords for MT4 platform or other needs

Description

Utility function to generate random passwords. Wrapper of cryptographic functions from 'openssl' library in R. Password length can be customized. By default function just output randomly generated 8 symbol password suitable for MT4 logins. It is also possible to create other passwords and include special symbols. When required, it's possible to write resulting password to the txt file. Once generated, password is written to the destination supplied by the user.

Usage

util_generate_password(
  salt = "something random",
  pass_len = 8,
  write_file = FALSE,
  file_name = "",
  special_symbols = FALSE
)

Arguments

Details

Passwords are generated using sha512 cryptographic function from openssl package. System date and user 'salt' is used to supply initial text for cryptographic function. Hashing function is using additional 'salt' which will be based on the current System time. Additionally, only a part of generated string is selected and used for password. Some letters of generated string are converted from lower to upper case.

Value

string or text file with password

Author(s)

(C) 2019, 2021 Vladimir Zhbanko

Examples

library(stringr)
library(magrittr)
library(openssl)
library(readr)

dir <- normalizePath(tempdir(),winslash = "/")
file_path <- file.path(dir, 'p.txt')

#write to file
util_generate_password(salt = 'random text', file_name = file_path)

#generate 8digit
util_generate_password(salt = 'random text')

#generate password with special symbols
util_generate_password(salt = 'random text', special_symbols = TRUE)

#generate longer password with special symbols
util_generate_password(salt = 'random text', pass_len = 10, special_symbols = TRUE)

Calculate Profit Factor

Description

Calculate profit factor using a data vector with the trading results

Usage

util_profit_factor(x)

Arguments

Value

function should calculate profit factor for this vector and return one value also as vector

Author(s)

(C) 2019 Vladimir Zhbanko

Examples

library(magrittr)
library(dplyr)
library(lazytrade)
data(profit_factor_data)
profit_factor_data %>%
   group_by(X1) %>%
   summarise(PnL = sum(X5),
             NumTrades = n(),
             PrFact = util_profit_factor(X5)) %>%
   select(PrFact) %>% head(1) %>% round(3)

Write csv files with indicated commands to the external system

Description

Function is capable to read the data and writing multiple files e.g. 'SystemControl8139124.csv'

Usage

write_command_via_csv(x, path_terminal = "", fileName = "SystemControl")

Arguments

Value

Function is writing multiple files e.g. 'SystemControl8139124.csv' to the Sandbox

typical content of the file: "Magic","IsEnabled" 8139124,1

Author(s)

(C) 2019, 2021 Vladimir Zhbanko

Examples

library(dplyr)
library(readr)
library(lubridate)
library(lazytrade)

path_sbxm <- normalizePath(tempdir(),winslash = "/")

file.copy(from = system.file("extdata", "OrdersResultsT1.csv", package = "lazytrade"),
          to = file.path(path_sbxm, "OrdersResultsT1.csv"), overwrite = TRUE)

DFT1 <- import_data(path_sbxm = path_sbxm,
                    trade_log_file = "OrdersResultsT1.csv")

DFT1 %>%
group_by(MagicNumber) %>% select(MagicNumber) %>% mutate(IsEnabled = 0) %>%
# head to shorten time of this example
head(2) %>%
# write commands to disable/enable systems
write_command_via_csv(path_terminal = path_sbxm)

Create initialization files to launch MT4 platform with specific configuration

Description

Function generate initialization files suitable for launching MT4 terminal with specific parameters. Several options available for generating files specific for each purpose. Option 'prod' will just use existing profile and connect to the broker server Option 'backtest' will generate file for the robot backtest Option 'opt' will generate file needed for the robot optimization Option 'full' allows to specify any desired parameter

Usage

write_ini_file(
  mt4_Profile = "Default",
  mt4_MarketWatch = "Forex.set",
  mt4_Login = "1234567",
  mt4_Password = "xxxxxXX",
  mt4_Server = "BrokerServerName",
  mt4_AutoConfiguration = "false",
  mt4_EnableNews = "false",
  mt4_ExpertsEnable = "true",
  mt4_ExpertsDllImport = "true",
  mt4_ExpertsExpImport = "true",
  mt4_ExpertsTrades = "true",
  mt4_Symbol = "EURUSD",
  mt4_Period = "H1",
  mt4_Template = "Default",
  mt4_Expert = "",
  mt4_ExpertParameters = "",
  mt4_Script = "",
  mt4_ScriptParameters = "",
  mt4_TestExpert = "",
  mt4_TestExpertParameters = "",
  mt4_TestSymbol = "EURUSD",
  mt4_TestPeriod = "H1",
  mt4_TestModel = "",
  mt4_TestSpread = "",
  mt4_TestOptimization = "false",
  mt4_TestDateEnable = "true",
  mt4_TestFromDate = "",
  mt4_TestToDate = "",
  mt4_TestReport = "test report",
  mt4_TestReplaceReport = "false",
  mt4_TestShutdownTerminal = "",
  mt4_TestVisualEnable = "false",
  dss_inifilepath = "",
  dss_inifilename = "test.ini",
  dss_mode = "prod"
)

Arguments

Details

added value of this function is the ability to generate multiple files to backtest several robots for several timeframes. For example it is solves the problem of doing repetitive tasks to 'backtest' robots for several currencies and repeat this procedure over time.

Most of the variables present in the function are starting with a prefix mt4_, the remainder of the name comes from the platform documentation, see references

Remaining variables are named with a prefix 'dss_' stands for 'Decision Support System', as these are the variables used for further automation purposes

Note that for simplicity reasons not all parameters are present in this function. e.g. FTP Settings and Proxy Server settings are not present

Value

output is a file with desired parameters

Author(s)

(C) 2019 Vladimir Zhbanko

References

All parameters used are taken from the reference documentation https://www.metatrader4.com/en/trading-platform/help/service/start_conf_file

Examples

library(lazytrade)

dir <- normalizePath(tempdir(),winslash = "/")

# test file to launch MT4 terminal with parameters
write_ini_file(mt4_Profile = "Default",
               mt4_Login = "12345678",
               mt4_Password = "password",
               mt4_Server = "BrokerServerName",
               dss_inifilepath = dir,
               dss_inifilename = "prod_T1.ini",
               dss_mode = "prod")

# test file to launch robot backtest
TO <- format(as.Date(Sys.Date()), "%Y.%m.%d")
FROM <- format(as.Date(Sys.Date()-60), "%Y.%m.%d")

# test file for MT4 use for backtesting
write_ini_file(mt4_Profile = "Default",
               mt4_Login = "12345678",
               mt4_Password = "password",
               mt4_Server = "BrokerServerName",
               mt4_TestExpert="FALCON_D\\Falcon_D",
               mt4_TestExpertParameters="Falcon_D.set",
               mt4_TestSymbol="EURUSD",
               mt4_TestPeriod="H1",
               mt4_TestModel="2",
               mt4_TestSpread="20",
               mt4_TestOptimization="false",
               mt4_TestDateEnable="true",
               mt4_TestFromDate=FROM,
               mt4_TestToDate=TO,
               mt4_TestReport="EURUSD_Report",
               mt4_TestReplaceReport="false",
               mt4_TestShutdownTerminal="true",
               mt4_TestVisualEnable="false",
               dss_inifilepath = dir,
               dss_inifilename = "backtest.ini",
               dss_mode = "backtest")

Table with a dataset to test the Model

Description

Table with a dataset to test the Model

Usage

x_test_model

Format

A dataframe with several columns

LABEL

future price change

X1-X75

Values of the macd indicator

Table with indicators and price change which is used to train model

Description

Table with indicators and price change which is used to train model

Usage

y

Format

A dataframe with several columns

X1

Time index

X2

Closed price now

X3

Closed price 34 bars ago

X4-X19

Series of Indicator values

LABEL

Price difference, difference between X3 and X2