The Naive (1/n) vs. Sharpe Portfolio: Out-of-Sample Performance Comparisons in Mid- and Small-Cap Equities

Recent research has demonstrated that many mean-variance portfolio selection rules often fail to out-perform the naive (1/n) portfolio in out-of-sample testing. This paper advances this line of inquiry by applying the naive and Sharpe Portfolios to 10000 data sets drawn from the 2007-2021 time frame. Unlike prior studies, however, the focus herein is on mid- and small-cap equities listed in the S&P 400 and S&P 600 indices, respectively. Additionally, both cross-sectional and time-series cross validation techniques are used to assess the out-of-sample performance. Shortfall performance is also assessed. The results indicate that the naive portfolio delivers inferior performance in-sample and superior performance out-of-sample, consistent with prior studies. This suggests that the naive model’s out-of-sample advantage persists in current data, across two very different modes of cross validation, and in equities with small- and mid-level market capitalization levels.