Testing Random Walk Model for Stock Daily Returns

Dr. Xinxin Jiang

Department of Mathematics and Computer Science
Rhodes College

February 22, 2005
202 Madison Hall
2:30 PM (Coffee and Cookies at 2:00)

Abstract

This talk will introduce a statistic that can be used to test independent and identically distributed (IID) random variables under reasonably weak assumptions. Its asymptotic normality can be shown under types of dependence (exchangeability and mixing stationary). One of its direct applications is to test whether the stock prices daily follow the classical random walk model. The model is rejected at any reasonable level of significance by using S&P 500 data from the 80's to date. The results will be discussed briefly.