Asymmetric Reaction to Information and Serial Dependence of Short-Run Returns

Abstract

This paper studies the daily stock price reaction to new information of portfolios grouped by size quintiles. To that end, cross-correlations, autocorrelations and Dimson beta regressions are analyzed. Based on a sample of shares traded in the Santiago de Chile Stock Exchange for the 1991-1998 period, results show that larger company stock prices - as measured by market capitalization - react to both good and bad news sooner than the smaller ones do. Thus a crossed effect appears, although not as a cascade: only the prices of large firms react earlier than the rest. These effects do not seem to be caused by non-trading. There also are significant asymmetric lagged and cross-effects. Good news has a more pronounced lagged effect than bad news does.

Keywords: Efficient market hypothesis, cross-serial autocorrelation, emerging markets

JEL Classification: G12, G15