Topic: Testing Normality of Financial Price Changes at High Frequencies

Speaker: George Tauchen, Professor of Economics and Finance, Duke University

Date: October 9, 2013 (Wed.)

Time: 1:00-2:30pm

Location: Building 4, Room 101

Language: English

Abstract:

The assumption that financial price changes are locally normally distributed is pervasive throughout finance and financial econometrics. The assumption is critical for both the economic and statistical theory. We derive formal tests of normality that are statistically valid. The idea is simple. We trim the price changes to remove large jumps and then scale by an estimate of the local volatility. We then justify the use of a Kolmogorov-Smirnov test for normality. The empirical application finds the normality assumption reasonable for the S&P stock price index but not valid for implied volatility of out-of-the-money options.

About the speaker:

George Tauchen is the William Henry Glasson Professor of Economics and Professor of Finance, Fuqua School of Business. He obtained his Ph.D. degree from the University of Minnesota in 1978 and BA degree from the University of Wisconsin in 1971.

His research interests focus on the Econometrics and Financial Economics. Dr. Tauchen is an internationally known time series econometrician. He has developed several important new techniques for making statistical inference from time series data and for testing models of financial markets.

He is co-editor of the Journal of Financial Econometrics, former editor of the Journal of Business and Economic Statistics, and former associate editor of Journal of Econometrics, Econometric Theory, Econometrica, Journal of Empirical Finance, and the Journal of the American Statistical Association.