Implementing stock selection ideas: Does tracking error optimization do any good?

Quantitative portfolio construction in active equity management requires forecasts on returns, covariances, and transaction costs. Optimizers use these forecasts as if they were certain. Given the error-maximizing nature of the optimization tool, uncertain forecasts may cause ex ante optimal portfolios to look ex post inefficient. The author offers an alternative portfolio construction technique called portfolio segmentation. It controls for tracking error risk by dividing the portfolio into an active and passive subportfolio. This approach does not require covariance forecasts and is considerably simpler to implement than the optimization technique. The portfolio segmentation technique has a good chance of doing at least as well as the optimization technique, unless the covariance forecasts for optimization are of high quality, in which case the optimization technique does better.