FIN 566 Fall 2020 Team 3 Final Project

Implementation of a High Frequency trading Strategy using the Strategy Studio backtesting software

Group info:

---

Khavya Chandrasekaran [email protected] - MSFE

Xuehui Chao [email protected] - MSFE

Yong Xie [email protected] - PhD Mathematics

Kushal Goenka [email protected] - MEng Electrical and Computer Engineering

---

Project Structure

• strategy_studio: contains strategy implementation in strategy studio
• trading_signals: trading signal analysis code for the strategy, mostly in Python.
• metrics: portfolio and strategy evaluation metric module.

Weekly Zoom Link Meeting:

---

Join Zoom Meeting https://illinois.zoom.us/j/89410527998?pwd=ZzJQbGpCb1NxY3N2dkk2Y1JCTkhxQT09

Meeting ID: 894 1052 7998 Password: 172526

How to run a Strategy

---

• First Build the required strategy via:

    cd /home/vagrant/Desktop/fin566_fall_2020_group_three/strategy_studio/localdev/RCM/StrategyStudio/examples/strategies/Team3Strategy
  
    make clean
  
    make all
  

• This results in a <"StrategyName">.so file.

    /home/vagrant/Desktop/strategy_studio/backtesting/./StrategyServerBacktesting
  

• Copy it over to the required folder cp /home/vagrant/Desktop/strategy_studio/RCM/StrategyStudio/examples/strategies/Team3Strategy/Team3Strategy.so /home/vagrant/Desktop/strategy_studio/backtesting/strategies_dlls

• Create an instance of the trading strategy

    create_instance Team3StrategyForSPY Team3Strategy UIUC SIM-1001-101 dlariviere 10000000 -symbols SPY
  

• Start the backtest

    start_backtest 2019-10-30 2019-10-30 Team3StrategyForSPY 1
  

• Now, find the local name of the most recently run backtest, by running

    ls -lA /home/vagrant/Desktop/strategy_studio/backtesting/backtesting-results
  

• To export the files and convert to CSV files:

    export_cra_file backtesting-results/BACK_SimpleTradePairsTest_2019-11-19_045116_start_10-30-2019_end_10-30-2019.cra backtesting-cra-exports
  

• The results will be in:

    cd /home/vagrant/Desktop/strategy_studio/backtesting/backtesting-cra-exports
  

Running a strategy with the automationScript

First set command line to DISABLE_COMMAND_LINE=false in backtester_config file.

Create an instnace of the strategy using piped multiple symbols:

create_instance S17 Team3Strategy UIUC SIM-1001-101 dlariviere 10000000 -symbols SPY|MSFT

Reset command line to DISABLE_COMMAND_LINE=true in backtester_config file.

Important Links:

https://davidl.web.engr.illinois.edu/data/20191030_marketdata/

https://davidl.web.engr.illinois.edu/fin566_fall_2020/lecture_videos/index.html

RCM Strategy Studio API Documentation: https://davidl.web.engr.illinois.edu/data/20191030_marketdata/rcm/RCM/StrategyStudio/docs/html/

RCM's Strategy Studio CSV input data format (if your project outputs in this format, you will later have an easier time backtesting on that data using strategy studio): https://davidl.web.engr.illinois.edu/data/20191030_marketdata/TextTickReader.pdf

Additional Strategy-Studio formatted ticks (trades) data decoded from IEX for most of month of October (2019-10-01 through 2019-10-30): link; Extract into the backtesting/text_tick_data folder using "tar -xjf ticks_201910.tar.bz2". This includes data for SPY,IVV,VOO,VXX,QQQ,TLT,LQD,HYG,USO,SPXL,DDG,FINZ,OILD,PSQ,PST,SDS,MSFT,AAPL,AMZN,FB,GOOGL,GOOG,DIA,BA,UNH,AAPL,GS,HD,MCD,V,MMM,UTX,DIS,CAT,JNJ,TRV,JPM,IBM,PG,AXP,WMT,CVX,NKE,MRK,XOM,VZ,WBA,INTC,KO,DOW,CSCO,PFE

https://davidl.web.engr.illinois.edu/data/20191030_marketdata/ticks_201910.tar.bz2