Skip to content

Latest commit

 

History

History
76 lines (58 loc) · 2.68 KB

README.md

File metadata and controls

76 lines (58 loc) · 2.68 KB
Raw

AutoTrade

This sample project is composed of current best practicies for building an automated trading application. Some of the features and functions represented:

  • parameters supplied by a configuration file
  • live chart being update with quotes, ticks, volume, and trade results
  • state machine to process entries and exits
  • portfolios, positions, instruments, orders and executions are recorded in a sqlite3 database
  • live data is sourced from IQFeed
  • orders are executed at Interactive Brokers (run this as a paper trading example)
  • trading is automatically limited to North America trading hours
  • quotes and ticks can be manually saved to an HDF5 based timeseries database at end of session

This project can be used as a template to try out your own high-frequency trading ideas.

Quotes are used to supply the trading signals in a high-frequency fashion.

x64/debug/AutoTrade.cfg

This example uses a three time period exponential moving average to generate the entry/exit signals.
The parameters are sourced in the configuration file (as an example):

x64/debug/AutoTrade.cfg:

ib_instance=2
symbol=SPY
#symbol=@ES~
# more market indicator combinations found  in ArmsIndex/PanelArmsIndex.cpp
symbol_tick=JT6T.Z
symbol_trin=RI6T.Z
period_width=5 #seconds
ma1_periods=5
ma2_periods=8
ma3_periods=12
stochastic_periods=17
group_directory=/app/AutoTrade/20220223 20:05:28.214488-1
sim_start=off

When there is no data available, set the group directory such nothing follows the '=', and set sim_start to off

Once timeseries data has been saved after a live session, it will be available in the TradeFrame.hdf5. The data can be queried by Hdf5Chart. The path can be copied in Hdf5Chart and pasted in to AutoTrade.db file for the 'group_directory=' entry. Simulation will then use this path for running simulations.

x64/debug/AutoTrade.db

The database has a number of tables, and can be accessed in a manner similar to this example:

$ sqlite3 x64/debug/AutoTrade.db
SQLite version 3.34.1 2021-01-20 14:10:07
Enter ".help" for usage hints.
sqlite> select * from portfolios;
Master|aoTF||1|1|USD|Master of all Portfolios|0.0|0.0
USD|aoTF|Master|2|1|USD|CurrencySummary of USD Portfolios|0.0|0.0
sqlite> select * from positions;
1|USD|SPY||ib01|iq01|SPY|ema|0|0|0|0|0.0|0.0|0.0|0.0
sqlite> .quit

For testing purposes, delete or rename the database to start fresh.

Simulation

NOTE: I have to test this yet

  • collect data during a live session, and at the end of the session, use the Save Values menu item
  • restart the application, and set D1 & X to Sim, and turn on
  • a Simulation menu will appear
  • Simulation->Run to start the simulation
    • trades should show up in similar time frames