Machine Learning Based Trading Strategies

Abstract

Dynamic contributions to trading are evaluated using covariations between position and price changes a horizon. Other performance measures like Sharpe ratios, Gain loss ratios, Acceptability indices and Drawdowns are also employed. Machine learning strategies based on Gaussian Process Regression (GPR) are compared with Least Squares (LSQ). Further both are generalized by invoking conservative valuation schemes that lead to the study of conservative conditional expectations modeled by distorted expectations. The latter lead to the development of distorted least squares (DLSQ) and distorted Gaussian Process Regression (DGPR) as the associated estimation or prediction schemes. Trading strategies are executed for nine sectors of the US economy using fourteen different predictive factor sets. Results support improvements made by GPR, DGPR over LSQ, DLSQ with the distorted versions also impacting favorably the drawdowns.

Keywords: Self Financing, Distorted Expectations, Active Portfolio Management, Zero covariation covariance

JEL Classification: G10, G12, G17