TFire

TFire is tool designed for finding and testing trading strategies on financial time series data. It combines the best properties of leading Python backtesting frameworks but is written in Julia. It is fully scriptable (in Julia) and extensible and accessed via a REPL.

If you want to dive right into it a recommended starting point is the Tutorial - Basic Analysis of Time Series Data.

Julia

TFire comes with the Julia language ecosystem. Anyone comfortable with scripting in Python will pick up scripting in Julia within a few days. The main advantage over Python for a backtesting application is speed and memory efficiency. While additions to Python like Numba, Cython and lately Mojo can deliver similar efficiency as Julia, none of them are as flexible and mature.

An introduction to Julia: https://mint.westdri.ca/julia/intro_intro

Installing and Running

Windows

  1. Download TFire folder for Windows and place it at your desired location.

  2. Navigate to the TFire folder and run run_tfire.bat

MacOS

  1. Download TFire folder for MacOS and place it at your desired location.

  2. Open a command prompt in the TFire folder and write ./run_tfire.sh

Exit TFire

To exit TFire write exit() at the prompt.