Allocation to industry portfolios under Markov switching returns

Abstract

Purpose – The purpose of this paper is to examine the effects of parameter uncertainty in the returns process with regime shifts on optimal portfolio choice over the long run for a static buy-and-hold investor who is investing in industry portfolios. Design/methodology/approach – This paper uses a Markov switching model to model returns on industry portfolios and propose a Gibbs sampling approach to take into account parameter uncertainty. This paper compares the results with a linear benchmark model and estimates without taking into account parameter uncertainty. This paper also checks the predictive power of additional predictive variables. Findings – Incorporating parameter uncertainty and taking into account the possible regime shifts in the returns process, this paper finds that such investors can allocate less in the long run to stocks. This holds true for both the NASDAQ portfolio and the individual high tech and manufacturing industry portfolios. Along with regime switching returns, this paper examines dividend yields and Treasury bill rates as potential predictor variables, and conclude that their predictive effect is minimal: the allocation to stocks in the long run is still generally smaller. Originality/value – Studying the effect of regime switching behavior in returns on the optimal portfolio choice with parameter uncertainty is our main contribution.