Essays in empirical finance

This dissertation consists of two essays. The first essay analyzes different properties of intraday returns. The second essay studies performance persistence of hedge fund returns.; In the first essay I check for normality of intraday returns after subordinating them using different trading activity measures. First, I define and discuss trading activity measures and explain the use of these measures. Second, I conclude that volume is as important as the number of transactions when measuring trading activity. Third, I examine specific features of intraday returns and show how to take them into account. Finally, I find that the normality of individual stocks (on the example of Intel and Cisco) cannot be restored easily. I show that after subordinating the returns by the volume process, the new return process still does not follow the normal distribution.; In the second essay I empirically demonstrate using data from Hedge Fund Research Asset Management (HFR) that both hot and cold hands among hedge fund managers tend to persist. While measuring performance I use statistical model selection methods for identifying style benchmarks for a given hedge fund and take into account the possibility that hedge fund net asset values may be computed using stale prices. I develop a new method for taking into account the backfill bias in the hedge fund database. I also take into account the selection bias introduced by the fact that hedge funds may go out of the HFR database. Using my method I estimate that a hedge fund that outperformed its style benchmark by 100 basis points in one year will on average outperform its style benchmark by 70 basis points the next year. In contrast a financial analyst who ignores the selection bias will find little persistence in performance.