Differential Interpretation of Information and the Post-Announcement Drift: A Story of Consensus Learning

55 Pages Posted: 23 Sep 2011

See all articles by António M.R.G. Barbosa

António M.R.G. Barbosa

ISCTE-IUL - Instituto Universitário de Lisboa

Date Written: September 23, 2011

Abstract

I show how a post-announcement drift can be generated in a model with fully rational investors who interpret public information differently. Differential interpretation of information transforms public raw information into private interpreted information. If investors recognize their limited ability to interpret information, they will look for other investors’ opinions in prices. Noise trading prevents investors from learning the market consensus interpretation of the announcement from the observation of a single price. But if noise trading follows a mean-reverting process, investors can gradually learn the market consensus from the observation of a series of prices. As investors become more confident about their interpretation of the announcement, they put more weight on it, and the announcement is gradually incorporated into prices, which generates a post-announcement drift. The model accounts for all salient empirical facts related to the post-announcement drift and delivers two new testable implications. If, in addition, investors make mistakes in extracting information from prices, the model also generates momentum.

Keywords: post-announcement drift, momentum, differential interpretation, learning, explanation

JEL Classification: G11, G12, G14

Suggested Citation

Barbosa, António M.R.G., Differential Interpretation of Information and the Post-Announcement Drift: A Story of Consensus Learning (September 23, 2011). Available at SSRN: https://ssrn.com/abstract=1932654 or http://dx.doi.org/10.2139/ssrn.1932654

António M.R.G. Barbosa (Contact Author)

ISCTE-IUL - Instituto Universitário de Lisboa ( email )

Av. das Forcas Armadas
Lisbon, 1649-026
Portugal
+351 21 790 39 16 (Phone)

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
92
Abstract Views
1,390
Rank
506,051
PlumX Metrics