Is Momentum Due to Data-snooping?

This paper explores the profitability of momentum strategies, by investigating if a momentum strategy is superior to a benchmark model once the effects of data-snooping have been accounted for. Two data sets are considered. The first set of data consists of US stocks and the second one consists of Swedish stocks. For the US data strong evidence is found of a momentum effect and hence the hypothesis of weak market efficiency is rejected. Splitting the sample in two parts, it is found that the overall significance is driven by events in the earlier part of the sample. The results for the Swedish data indicate that momentum strategies based on individual stocks generate significant profits. A very weak or no momentum effect can be found when stocks are sorted into portfolios. Finally, and perhaps most importantly, results show that data-snooping bias can be very substantial. Neglecting the problem would lead to very different conclusions.