KrakenBacktesting
Backtesting for cyrpto trading strategies on Kraken
What does this do? Why was this made?
The intentions of this project was to create an enviroment that simulates the order books on cryptocurrency markets; I chose Kraken as the market of choice since it was easy enough to pull all of the historical data from a particular trading pair, some markets dont offer this. What makes this testing library different from other backtesting financial libraries is that it was intended to be used for cyrpocurrency markets; so instead of relying on particular candle intervals, the precision can go all the way down to a per orderbook update logic; so you can make the update interval nearly anything you like. It also has different functionality when it comes to placing orders or stop limits; no assumptions are made on the buying or selling of crypto currency, meaning someone must fill your order for it to go through.
Currently the only functionality of this library is to show the wallet's balances and the total balance of the wallet as it runs through the interval of data given to it; so it is currently only good to see if a strategy could be successful, any graphing or analytical pieces simply havent been developed yet but are being planned.
Setup
To run this script you will need
- python 3.4
- Pandas
- MatPlotLib
- TA-Lib
How do I use this?
Included in this project is a file titled 'kraken_rig_runner' which I have been using for testing purposes and should be easy enough adapt and follow; I will write a more in depth guide on usage in the future.
to-do
- setup.py
- plotting
- archiving results for future analysis
- adding basic analytical functions
- fix precision error
- clean and refine code already in place