Diurnal rhythms in investor sentiment

I use data from a large social network to assess the diurnal stability of investors' collective valuation of financial assets. Employing user-assigned valuation indicators attached to time-stamped messages, I show that investors' sentiment towards assets, which is fairly constant during most of the day, dips markedly and regularly in the morning. When looking at messages posted by different subsamples of users, I find that both level and variability of investor sentiment decrease with trading experience. Not only are more experienced investors as a group less optimistic than novice investors, their valuations are also substantially less variable over the course of the day. The findings provide empirical support for assumptions and results that feature prominently in theoretical and empirical descriptions of investor behavior. (C) 2015 Elsevier B.V. All rights reserved.