Solutions A T-Stat of 2 As discussed briefly in Step 3, one solution to determine manager skill is to identify if there are enough years of performance data to be statistically significant by measuing a manager’s t-stat. If the t-stat is 2 or greater, then the investor has at least a 95% confidence level that the manager’s above-benchmark returns were due to skill, with up to a 5% chance that they were due to luck. The four “alpha charts” (Figures 5-11 through 5-14) show the year-by-year difference between the fund return and the benchmark return for four funds with at least ten years of data whose managers who were recognized by Morningstar as “Manager of the Year” for 2013. While all of these funds had a positive average alpha, none of them had a t-stat of 2 or greater, which means that their alpha was not consistent enough to be deemed a function of skill rather than luck. Among the four managers shown to have positive alpha, the average minimum number of years to determine luck vs. skill was 47.5, with the most consistent fund having a minimum track record of 20 years (Figure 5-13) and the least consistent having a minimum track record of 118 years (Figure 5-14). When managers are subjected to the scrutiny of a simple t-test, the idea of manager skill to produce consistent alpha quickly becomes relegated to the realm of fantasy, taking its rightful place alongside unicorns, Bigfoot, and the Loch Ness Monster - as seen depicted in The Alpha Myth painting on the following page.
Published on Jun 1, 2015
This book reveals the potential land mines and pitfalls of active investing and educates readers on the benefits of passive investing with i...