/pdbox

Loop-based pandas backtester

Primary LanguagePython

This project is an algorithm backtesting framework which is still on the works

It uses a loop based approach, which is faster but may produce biased results, so use carefully.

This project uses:

  • pandas
  • numpy
  • requests
  • futures

The project has not been in development since 2017.

Project has been picked up Again in Jul 13.

Requirements

  • Python 2.7

TODO

  • Migrate to Python 3.7

  • Refactor as a modularized approach

  • Refactor for best practices

  • Refactor for best software design patterns

  • Use formatter

  • Use linting

  • Use Precommit hooks

  • Improve README

  • Add tests

  • Add documentation

  • Add static types

  • Improve performance

    tickers = [

    'DPW', 'GROW', 'SGRP', 'CBIO', 'GBR', 'MOSY', 'CREG', 'ONCS', 'EVEP',

    'ISIG', 'CALI'

    ]

    #tickers = ['GBR'] #tickers = ['SRAX','KOOL','DYSL']

    tickers = ['AAPL, MSFT, TWTR']

    tickers = ['ONP', 'CNET', 'RCON', 'CREG']