IBrokers2: R API to Interactive Brokers (clone of IBrokers)

Create empty package directory with the R command: Rcpp::Rcpp.package.skeleton()
In the R sub-directory of the new package, create a file called my_package.R and add the header: #' @useDynLib my_package #' @importFrom Rcpp evalCpp #' @exportPattern "^[[:alpha:]]+" Add your R functions below the above header.
In the Build menu, Clean and rebuild

Install package from source on local drive install.packages(pkgs="C:/Develop/R/IBrokers2", repos=NULL, type="source") Install package from source on local drive using R CMD R CMD INSTALL C:\Develop\R\IBrokers2 Install package from github devtools::install_github(repo="algoquant/IBrokers2", force=TRUE) Build vignette package reference manual from *.Rd files system("R CMD Rd2pdf C:/Develop/R/IBrokers2") cd C:\Develop\R\IBrokers2\vignettes R CMD Rd2pdf C:\Develop\R\IBrokers2\

The function .trade_realtime() performs a loop requesting real time bars for all instruments.
The parameter trade_params is passed into the function e_wrapper$model_fun(), and it isn't used outside it.
Currently the design of e_wrapper$realtimeBars() only allows to trade a single instrument at a time.
The numeric IB API message codes are stored as named lists: IBrokers2::.twsIncomingMSG IBrokers2::.twsOutgoingMSG
In e_wrapper$realtimeBars(), when would this ever be true: is.null(trade_params) ? Why is it needed? Answer: It is true for instruments which aren't traded, because they have trade_params equal to NULL

[ ] How to open IB demo account https://www.interactivebrokers.com/en/index.php?f=1286
[ ] How to trade IB using Python configure IB account https://medium.com/auquan/algorithmic-trading-system-development-1a5a200af260

[ ] Create vignette for package IBrokers2
[ ] Add software disclaimer
[ ] Add model for ES strategy trading using VX curve by executing market orders
[ ] Fix pairs_strat() in trade_wrapper.R: align the bar data according to the date stamps, then fix calculation of z-score The traded instrument should be second in the list so that the model is run on contemporaneous data, after the bars for both instruments arrive?
[x] In realtimeBars() calculate the trailing VWAP and volatilities
[ ] Modify the function e_wrapper$realtimeBars() to be able to trade several instruments simultaneously The function e_wrapper$model_fun() would require access to the trade_params parameter of several instruments.
[ ] In realtimeBars() check for trade status using reqOpenOrders() instead of copying tradeID https://stackoverflow.com/questions/34703679/r-ibrokers-reqopenorders-hangs
[ ] In create_ewrapper() modify the handlers openOrder() and openOrderEnd() for reqExecutions() and reqOpenOrders()
[ ] In IB_scripts.R use limit order with GoodTillDate order expiration, instead of canceling the order https://interactivebrokers.github.io/tws-api/classIBApi_1_1Order.html#a95539081751afb9980f4c6bd1655a6ba
[ ] Create a clone of reqOpenOrders() called get_open_orders() to write to data buffer - reqOpenOrders() doesn't work, just runs in endless loop
[ ] Find out how to book iBrokers trades in different models (portfolios) https://www.interactivebrokers.com/en/software/tws/usersguidebook/modelportfolios/createmodel.htm
[ ] Adapt from IB script using R6Class: C:/Develop/R/IBrokers2/scripts/TWS Kovalevsky.R
[ ] Demonstrate how to use iBrokers data playback replay feature The script added to data_management.Rnw downloads the raw data, but doesn't replay the bar data properly. https://offerm.wordpress.com/2015/05/21/market-data-recording-and-playback-with-ibrokers-and-r-2/
[ ] Adapt from: downloading raw market data using IBrokers::reqMktData() without eventWrapper https://stat.ethz.ch/pipermail/r-sig-finance/2011q3/008232.html
[ ] Print to console status of the eWrapper data buffer
[ ] In create_ewrapper() modify openOrder() to write to the da_ta environment and to a file
[ ] Create a clone of reqAccountUpdates() called get_account()
[ ] Create sub-portfolios and place trades into sub-portfolios: use modelCode ? https://www.interactivebrokers.com/en/software/tws/usersguidebook/mosaic/portfoliobuilder.htm interactivebrokers sub portfolios
[ ] Create a shiny app as a front end for trading via IBrokers2: C:\Develop\R\IBrokers2\scripts\app_ibtrading.R Every time model parameters are updated, the shiny app should interrupt the trading model in R and then restart it with the new parameters. Use on.exit() to remember the trade IDs. I've been trying to find a solution for this problem, but only stumbled into vague mentions. The goal is to receive live data from a trading platform with IBrokers package, accumulate it, do computations with it periodically and as the user changes reactive inputs. The package uses subscribe-callback mechanism to process new data. Once the data is requested, the function goes to a while(TRUE) loop and passes incoming messages to a callback function, where it can be written to a data frame, until stopped. Although it is not computationally intensive, it occupies the session. Is it possible to update the data on a background and periodically do the calculations with it? I suspect it somehow involves multiple R sessions, since R is single threaded. I would appreciate any tips. https://groups.google.com/forum/#!topic/shiny-discuss/n11-mnBYXQc https://github.com/ksavin/intrinio https://www.linkedin.com/in/ksavin/ https://gist.github.com/trestletech/8608815 https://stackoverflow.com/questions/21282228/update-plot-within-observer-loop-in-shiny-application
[ ] Remove Depends: and Imports: from DESCRIPTION ?
[ ] Modify IBrokers2 startup message in zzz.R
[ ] Add sounds when trades are placed https://stackoverflow.com/questions/3365657/is-there-a-way-to-make-r-beep-play-a-sound-at-the-end-of-a-script https://shirinsplayground.netlify.com/2018/06/text_to_speech_r/ https://sourceforge.net/projects/espeak/files/espeak/ http://code.markedmondson.me/googleLanguageR/articles/text-to-speech.html https://github.com/seankross/ari cd C:\Program Files (x86)\espeak\command_line espeak.exe -v english-us -s 100 "Buy"
[ ] Every 10 counts, save all the bar_data, instead of every row, or save only 10 rows of bar_data at a time to a file, instead of every row.
[ ] Download messages and parse them with reqMktData(tws, equity1, CALLBACK=NULL, file=fn) http://r.789695.n4.nabble.com/Interactive-Brokers-td3668170.html
[ ] Create new order type: https://stackoverflow.com/questions/46482300/r-ibrokers-interactive-brokers-api
[ ] Run reqExecutions(): https://stackoverflow.com/questions/35559742/reqexecutions-ibrokers-package http://r.789695.n4.nabble.com/IBrokers-order-status-td4562387.html http://r.789695.n4.nabble.com/Confirm-orders-in-TWS-Interactive-Brokers-td3785506.html
[ ] Adapt iBrokers code from: https://www.quantopian.com/posts/ib-api https://stackoverflow.com/questions/35559742/reqexecutions-ibrokers-package http://r.789695.n4.nabble.com/Confirm-orders-in-TWS-Interactive-Brokers-td3785506.html
[ ] Answer the questions: https://stackoverflow.com/questions/39713543/message-management-using-ewrapper-and-twscallback-in-ibrokers https://stackoverflow.com/questions/27254131/ibrokers-queueing-up-an-order-using-r
[ ] Reply to / comment on: https://stackoverflow.com/questions/44582053/ibrokers-quantstrat-live-implementation
[ ] Answer the question: https://stackoverflow.com/questions/21442158/ibrokers-data-persistence
[ ] Add new order type to IBrokers: https://stackoverflow.com/questions/46482300/r-ibrokers-interactive-brokers-api
[ ] Adapt from autotrade by Phillip Guerra: uses package reticulate with Python for Interactive Brokers https://www.pcppresentation.com/ https://github.com/PhilGuerra/autotrade https://github.com/PhilGuerra/autotrade/commits?author=PhilGuerra Guerra autotrade RFinance Chicago 2018.pdf C:/Research/R/Tutorials/Guerra_automated_trading_with_r
[ ] Adapt from Santosh Srinivas: Interactive Brokers with Python, R and package reticulate https://www.santoshsrinivas.com/amicable-interactive-data-analysis-using-pythoth-and-r-using-reticulate/

[x] In realtimeBars() download from IB portfolio information: net positions, pnls
[x] In e_wrapper$realtimeBars(), call the model function using indirection with the name of the model function from the list trade_params Add the name of the model function to the list trade_params.
[x] In trade_wrapper() add limits on position inventory: if inventory reaches its limit then stop placing orders in that direction Using e_wrapper$da_ta$position[contract_id]. Add position limit to the list trade_params.
[x] Remove old file eWrapper_trading.R (old version of trade_wrapper.R)
[x] In e_wrapper$realtimeBars(), update EWMA and volatility for all instruments, even for non-traded instruments
[x] Set trade_params to NULL for non-traded instruments, instead of NA
[x] Fix bug in e_wrapper$realtimeBars: catch case when ib_account is NULL, for example at initial state when there are no positions
[x] In trade_wrapper(), rename variable sprea_d to bia_s
[x] Pass ac_count argument with acctCode into trade_wrapper()
[x] Call reqAccountUpdates() inside realtimeBars() to download net positions from IB
[x] In realtimeBars() and model_fun() add argument ib_connect In processMsg() pass argument twsconn into eWrapper$realtimeBars()
[x] In trade_realtime() rename the argument playback to back_test (with default FALSE), for backtesting trading strategies using 5-second bar data
[x] Create a clone of twsCALLBACK() called call_back() - rename argument playback to back_test, with default FALSE
[x] Move argument lamb_da to argument vector trade_params
[x] In trade_wrapper() add spread bias to the limit prices if there is momentum For example, use EWMA crossover. Or if there are two consecutive trades in the same direction.
[x] In trade_wrapper() add argument warm_up for warmup period: don't trade in warmup period Rename argument fac_tor to warm_up.
[x] Add argument trade_params to model_fun()
[x] In trade_wrapper() get rid of limit_prices
[x] In trade_wrapper() add ability to trade with a time lag Create a matrix of past limit_prices, and update the limit_prices in model_fun(). Set the current limit prices to past limit_prices.
[x] Rename limit_prices to trade_params, and add lagg parameter to trade_params
[x] Increase TWS Java heap size to 2.5 GB in file C:\Jts\tws.vmoptions -Xmx2500m https://www.interactivebrokers.com/en/software/tws/usersguidebook/priceriskanalytics/custommemory.htm https://ibkr.info/article/2170
[x] In trade_wrapper() add spread factor for limit price, proportional to the trailing volatility
[x] Create clone of reqRealTimeBars() called trade_realtime()
[x] In create_ewrapper() rename .Data to da_ta
[x] Use assign() instead of "<<-" - actually revert back to "<<-"
[x] In trade_realtime() rename con to sock_et, and conn to ib_connect, and twsconn to ib_connect
[x] In create_ewrapper() rename assign.Data() to as_sign()
[x] Pass con_tracts and limit_prices arguments into trade_wrapper(), and pass contract_id argument into model_fun() Trade only those contracts which have non-NA limit_prices.
[x] Download data for two contracts simultaneously
[x] Change the trading frequency - realtimeBars query in IB API is fixed to 5 seconds
[x] In realtimeBars() add to col_index the instrument ID column
[x] Pass into IBrokers2::reqRealTimeBars() a named list of contracts called con_tracts (even for a single contract)
[x] In trade_wrapper(), pass a vector of contract (instrument) names called name_s, instead of the integer n_instr Add column names to bar_data. Add argument file_connects, and write the column names as headers to data files. Remove the dots argument of trade_wrapper().
[x] Download real-time bars for multiple instruments using reqRealTimeBars() and eWrapper.RealTimeBars.CSV(), and compare the data
[x] In realtimeBars() of trade_wrapper(): copy bar data into buffer in the event wrapper environment
[x] Convert bar_data into list of matrices or data frames, instead of a list of xts series
[x] Rewrite trade_wrapper() Rename Wrapper_new() to create_ewrapper(). Remove redundant accessor function calls. Define model function inside trade_wrapper() and pass model parameters into trade_wrapper() through the dots argument.
[x] Create IBrokers2 project directory in RStudio
[x] Pass the trading parameters buy_spread and sell_spread into the dots of realtimeBars()
[x] Adapt Interactive Brokers API script from: actually nothing there to adapt C:\Develop\R\IBrokers\Sherrington IBrokers scripts.pdf C:\Develop\R\IBrokers\Yadav IBrokers scripts.pdf