Machine Learning Quant Strategies using clustering to identify similar risk and return characteristics.
- Unsupervised Learning Trading Strategy -
- Download/Load NIFTY50 stocks prices data.
- Calculate different features and indicators on each stock.
- Calculate Monthly Returns for different time-horizons.
- Download Fama-French Factors and Calculate Rolling Factor Betas.
- For each month fit a Clustering Algorithm to group similar assets based on their features.
- For each month select assets based on the cluster and form a portfolio based on Efficient Frontier max sharpe ratio optimization.
- Visualize Portfolio returns and compare to NIFTY50 returns.