An Alternative Behavioral Explanation for the MAX Effect

58 Pages Posted: 4 Aug 2018 Last revised: 1 Aug 2019

See all articles by Maren Baars

Maren Baars

University of Muenster - Finance Center

Hannes Mohrschladt

University of Muenster - Finance Center

Date Written: April 4, 2019

Abstract

Stocks with high maximum daily returns in the previous month (MAX) yield low future returns. We examine the underlying sources of this MAX effect and present three empirical arguments that question the common presumption that investors with lottery preferences cause an overvaluation of high-MAX stocks. First, high-MAX stocks are comparably unattractive for investors with cumulative prospect theory preferences. Second, we find no price pressure from lottery investors after high-MAX observations but immediate price reversals. Hence, the MAX return itself seems to be the source of the overvaluation. Third, the MAX effect reverses if the MAX return can be linked to an earnings announcement. These findings are perfectly in line with a behavioral phenomenon called strength-weight bias: Investors usually overreact towards extreme high-strength news such that high-MAX stocks tend to be overvalued. However, they under-react if the MAX return is accompanied by reliable high-weight news such as earnings announcements.

Keywords: MAX Effect, Lottery Preferences, Strength-Weight Bias, Cross-Sectional Return Predictability

JEL Classification: G12, G14, G40

Suggested Citation

Baars, Maren and Mohrschladt, Hannes, An Alternative Behavioral Explanation for the MAX Effect (April 4, 2019). Available at SSRN: https://ssrn.com/abstract=3214438 or http://dx.doi.org/10.2139/ssrn.3214438

Maren Baars

University of Muenster - Finance Center ( email )

Universitätsstraße 14-16
Münster, 48143
Germany

Hannes Mohrschladt (Contact Author)

University of Muenster - Finance Center ( email )

Universitätsstr. 14-16
Muenster, 48143
Germany

HOME PAGE: http://www.wiwi.uni-muenster.de/fcm/en/the-fcm/lsf/team/hannes-mohrschladt

Here is the Coronavirus
related research on SSRN

Paper statistics

Downloads
349
Abstract Views
1,338
rank
88,591
PlumX Metrics