CMStatistics 2016: Start Registration
View Submission - CFE
A1372
Title: Testing out-of-sample portfolio performance Authors:  Ekaterina Kazak - University of Manchetser (United Kingdom)
Winfried Pohlmeier - University of Konstanz (Germany) [presenting]
Abstract: While the literature on portfolio choice largely concentrates on stabilization strategies, little attention has been devoted to the quality of performance tests used to check, if a portfolio strategy can significantly outperform an alternative one in terms of a given performance measure. The quality of portfolio performance tests based on out-of-sample returns are studied. We argue that the puzzling empirical results of inferior performance of the theoretically superior strategies based on the out-of-sample comparison are partly resulting from the low power properties of these tests. We emphasize the importance of the underlying return distribution and show that the out-of-sample portfolio returns follow a mixture distribution depending on the return vector, but also the estimated portfolio weights. In our simulation study with the proposed mixture distribution design we show that in the realistic cases the test difference is overemphasized. The main issue is the low testing power, which automatically leads to a conclusion, that the benchmark strategy cannot be outperformed. For the applied researcher we provide some guidance to cope with the problem of low power.