Algorithmic Portfolio Tilting to Harvest Higher Moment Gains

Abstract

Many financial portfolios are optimized without taking the higher moments into account. We recommend tilting these portfolios in a direction that increases their estimated mean and third central moment and decreases their variance and fourth central moment. The advantages of tilting come at the cost of deviation from the initial optimality criterion. In this paper, we show the usefulness of portfolio tilting applied to the equally-weighted, equal-risk-contribution and maximum diversification portfolios in a UCITS-compliant asset allocation setting.

Keywords: mean-variance-skewness-kurtosis, non-normality, portfolio allocation, tilting

JEL Classification: G110, C100