Rebalancing index tracking portfolios with cumulative sum (CUSUM) control charts

In this study, we use the cumulative sum (CUSUM) control chart methodology to regulate index tracking portfolio updates over time, as we seek to make the rebalancing decision endogenous to the portfolio selection problem. We use data from two stock markets (United States and Brazil), and we estimate CUSUM based portfolios as well as portfolios using fixed rebalancing time windows. We also provide a comparison with the exponentially weighted moving average (EWMA) control chart methodology. Our findings show the suitability of CUSUM, in a dynamic condition in which we have more portfolio updates when tracking performance is poor (usually during periods when markets have more volatility) and lower updates when tracking performance is effective.