Assessing out-of-sample performance of orthogonal portfolio rules in emerging markets

Abstract

Orthogonal portfolios are valuable for enhancing out-of-sample performance in the presence of Gaussian returns. This study examines the effectiveness of implementable portfolio rules based on this approach when returns deviate from normality, and reflect the statistical properties of emerging market returns. Our findings indicate that the theoretical out-of-sample performance ranking of certain orthogonal rules, called implementable Q-rules, is not only valid for Gaussian data, but also for the non-Gaussian empirical distribution observed in our ETF dataset. © 2025 Elsevier Inc.