The purpose of this paper is to test the performance of momentum strategies and identify the sources of their profits.
To identify the main source of momentum profits, first, the bootstrap method with replacement was used. Then, to eliminate the existence of the small sample bias, the bootstrap method without replacement and the block bootstrap method were employed. In this case, when the authors draw the observations without replacement the random effect is reduced, whereas the resampling procedure is based on the random draw.
The empirical results show the existence of a small sample bias in the bootstrap method with replacement, and that the time‐series relations of stock returns are the main source of momentum profits.
To ensure the random effect of the draws, the authors develop a new resampling procedure called the mixed bootstrap method.
