We show that results in the recent strand of the literature, which tries to explain stock returns by weather induced mood shifts of investors, might be data-driven inference. More specifically, we consider two recent studies [Kamstra, Mark J., Kramer, Lisa A., Levi, Maurice D., 2003a. Winter blues: A SAD stock market cycle. American Economic Review 93(1), 324–343; Cao, Melanie, Wei, Jason, 2005. Stock market returns: A note on temperature anomaly. Journal of Banking and Finance 29(6), 1559–1573] that claim that a seasonal anomaly in stock returns is caused by mood changes of investors due to lack of daylight and temperature variations, respectively. While we confirm earlier results in the literature that there is indeed a strong seasonal effect in stock returns in many countries: stock market returns tend to be significantly lower during summer and fall months than during winter and spring months as documented by Bouman and Jacobsen [Bouman, Sven, Jacobsen, Ben, 2002. The Halloween indicator, Sell in May and go away: Another puzzle. American Economic Review, 92(5), 1618–1635], there is little evidence in favor of a SAD or temperature explanation. In fact, we find that a simple winter/summer dummy best describes this seasonality. Our results suggest that without any further evidence the correlation between weather-related variables and stock returns might be spurious and the conclusion that weather affects stock returns through mood changes of investors is premature.

Changes in oil prices predict stock market returns worldwide. We find significant predictability in both developed and emerging markets. These results cannot be explained by time-varying risk premia as oil price changes also significantly predict negative excess returns. Investors seem to underreact to information in the price of oil. A rise in oil prices drastically lowers future stock returns. Consistent with the hypothesis of a delayed reaction by investors, the relation between monthly stock returns and lagged monthly oil price changes strengthens once we introduce lags of several trading days between monthly stock returns and lagged monthly oil price changes.