Quantile Regression Based Enhanced Indexing with Portfolio Rebalancing

In this article, an enhanced indexing (EI) model is put forward for portfolio selection to optimize a specific quantile of the return distribution of the benchmark index + alpha return combined with the conditional value at risk to control the downside risk in the portfolio. Constraints on short-selling and portfolio rebalancing with the transaction and holding costs are built in the models as a means of integrating real-life functionalities. The proposed models are linear or mixed integer linear programs. The out-of-sample performance comparison of the proposed model without short-selling with the benchmark index and an existing quantile based EI model on the data sets of several markets across the globe exhibit higher average returns and mean-risk ratios like Sharpe ratio and STARR, thus fulfilling the objective of EI. We also study the out-of-sample performance of our model with short-selling during financial meltdown periods.