If you are familiar with ML, you may agree that feature engineering tends to be the most important ingredient for improving predictions. The same holds true for trading with a twist: on the one hand, there is decades of research into factors that drive asset returns and quantitative methods to predict prices; on the other hand, new data sources and ML create new opportunities and require re-evaluating established perspectives.
This category focuses on topics, code examples and new ideas related to the chapter Financial Feature Engineering: How to research Alpha Factors, including:
- Which categories of factors exist, why they work, and how to measure them,
- Creating alpha factors using NumPy, pandas, and TA-Lib,
- How to de-noise data using wavelets and the Kalman filter,
- Using Zipline to test individual and multiple alpha factors,
- How to use Alphalens to evaluate predictive performance.