Overfitting Prevention
Methodology
Overfitting Prevention involves splitting historical data into rolling windows.
Common Pitfalls
Avoid: survivorship bias (only analyzing surviving assets).
Validation
Perform out-of-sample testing on most recent 6 months.
Note: This documentation reflects testing on Ubuntu 22.04 with Python 3.10. Adjust paths and commands for your environment.
Further Reading
- FreqTrade official documentation:
freqtrade.io/en/stable - CCXT exchange library:
github.com/ccxt/ccxt - TA-Lib indicators:
ta-lib.org