The Impact of Microstructure Noise on the Distributional Properties of Daily Stock Returns.pdf

The Impact of Microstructure Noise on the Distributional Properties of Daily Stock Returns Standardized by Realized Volatility Jeff Fleming, Bradley S. Paye Jones Graduate School of Management, Rice University Abstract of the literature which directly tests for jumps. Andersen, Benzoni and Lund (2002) and Chernov, Gallant, Ghysels Previous studies find that daily stock returns standard- and Tauchen (2003), for example, find evidence of jumps ized by realized volatility are approximately standard based on parametric estimation of jump diffusion models, normal. This evidence suggests that jumps are not an and Andersen, Bollerslev and Diebold (2005), Huang and empirically relevant feature of stock prices, which is in- Tauchen (2005), Jiang and Oomen (2005), Tauchen and consistent with a growing body of research that directly Zhou (2005) and Barndorff-Nielsen and Shephard (2005, tests for and finds evidence of jumps. This paper re- 2006) find evidence of jumps using nonparametric tech- solves the apparent contradiction. We show that upward niques that exploit the information in high-frequency in- bias in realized volatility estimates due to microstructure traday returns. noise can artificially reduce the variance and increase the This paper resolves the apparent empirical contradic- kurtosis of standardized returns and lead to the false ap- tion. The essence of our argument is that the distribution pearance that returns are approximately standard nor- of standardized returns constructed using the standard mal. Using a