This project is a production ready prototype model that
- builds a cross-sectional statistical risk model using PCA, which is used to build a portfolio along with a number of momentum, mean reversion, and sentiment alpha factors
- universe: US equities, top 500 in terms of dollar volume
- evaluates the alpha factors, combining the promising signals into a single alpha vector
- evaluates a number of approaches to constructing the optimal portfolio given objective/constraints
note: uses end of day pricing data from Quotemedia and sector data from Sharadar, which could not be shared due to licensing restrictions