Anecdotal evidence and recent theoretical models argue that past stock returns affect subsequent stock trading volume. We study 3,000 individual investors over a 51 month period to test this prediction using linear panel regressions as well as negative binomial panel regressions and Logit panel regressions. We find that both past market returns as well as past portfolio returns affect trading activity of individual investors (as measured by stock portfolio turnover, the number of stock transactions, and the probability to trade stocks in a given month) and are thus able to confirm predictions of overconfidence models. However, contrary to intuition, the effect of market returns on subsequent trading volume is stronger for the whole group of investors. Using survey data of our investor sample, we present evidence that individual investors, on average, are unable to give a correct estimate of their own past realized stock portfolio performance. The correlation between return estimates and past realized returns is insignificant. For the subgroup of respondents, we are able to analyze the link between the ability to correctly estimate the past realized stock portfolio performance on the one hand and the dependence of trading volume on past returns on the other hand. We find that for the subgroup of investors that is better able to estimate the own past realized stock portfolio performance, the effect of past portfolio returns on trading volume is stronger. We argue that this finding might explain our results concerning the relation between past returns and subsequent trading volume.
