A Generalization of the Calendar Time Portfolio Approach and the Performance of Private Investors

We present a regression-based generalization of the calendar time portfolio approach which allows for decomposing the risk-adjusted performance of private investors (or firms or mutual funds) into multivariate and continuous subject characteristics. Our technique remedies several well-known weaknesses of the traditional calendar time portfolio approach and it ensures that the statistical results are heteroscedasticity consistent and robust to general forms of cross-sectional and temporal dependence. Considering a new, unique dataset on more than 40,000 European private investors, we validate some of the most popular hypotheses on the performance of private investors. Likewise, we show that erroneously ignoring cross-sectional dependence inherent in microeconometric panel data can lead to severely biased statistical results.