Market Timing For active Asset Allocation: A Discrete Regression Model Approach

Abstract

A discrete regression model (DRM) approach to timing the asset class markets for stocks, bonds, and cash in active asset allocation is presented. The technique involves investing in the asset class whose return is predicted to exceed the other asset class return after observing a sequential signal of estimated probabilities. The empirical results show that the DRM approach provides enhanced portfolio performance when compared to more passive fixed-weight portfolio strategies.