Online Learning of Portfolio Ensembles with Sector Exposure Regularization

TL;DR

This paper introduces an online learning method for portfolio ensembles that regularizes risk by promoting diversification across predefined sectors, achieving low regret and improved risk-adjusted returns.

Contribution

It proposes a novel online convex optimization approach that encourages diversification in portfolio ensembles based on sector groupings, with proven regret bounds.

Findings

Logarithmic regret with respect to the best ensemble in hindsight

Significant increase in Sharpe ratio in empirical tests

Effective regularization of risk through sector diversification

Abstract

We consider online learning of ensembles of portfolio selection algorithms and aim to regularize risk by encouraging diversification with respect to a predefined risk-driven grouping of stocks. Our procedure uses online convex optimization to control capital allocation to underlying investment algorithms while encouraging non-sparsity over the given grouping. We prove a logarithmic regret for this procedure with respect to the best-in-hindsight ensemble. We applied the procedure with known mean-reversion portfolio selection algorithms using the standard GICS industry sector grouping. Empirical Experimental results showed an impressive percentage increase of risk-adjusted return (Sharpe ratio).