2010•Munich Personal RePEc Archive (Ludwig Maximilian University of Munich)Requires access

Overconfidence, Risk Aversion and Individual Financial Decisions in Experimental Asset Markets

Julija Michailova

Open publisher page 1 citations

Abstract

We investigate the influence of overconfidence and risk aversion on individual financial decision making in the experimental asset markets of the Smith, Suchanek and Williams (1988) type, with no informational asymmetries. Subjects, based on their pre-experimental overconfidence scores, were assigned to the two types of markets: least overconfident subjects formed five “rational” markets and most overconfident subjects formed five “overconfident” markets. The asset market experiment was followed by post hoc risk aversion measurement. Our results revealed that in the suggested setting, performance and trading activity were overconfidence dependent only for female participants. Mistakes in price forecasting, that are negatively correlated with overconfidence, could partially account for the increase in trading activity and losses. In the decreased sample differences in individual outcomes were overconfidence and not risk aversion driven.

Open-access reader

About this research paper

What this paper is about

We investigate the influence of overconfidence and risk aversion on individual financial decision making in the experimental asset markets of the Smith, Suchanek and Williams (1988) type, with no informational asymmetries. Subjects, based on their pre-experimental overconfidence scores, were assigned to the two types of markets: least overconfident subjects formed five “rational” markets and most overconfident subjects formed five “overconfident” markets. The asset market experiment was followed by post hoc risk aversion measurement. Our results revealed that in the suggested setting, performance and trading activity were overconfidence dependent only for female participants. Mistakes in price forecasting, that are negatively correlated with overconfidence, could partially account for the increase in trading activity and losses. In the decreased sample differences in individual outcomes were overconfidence and not risk aversion driven.

Why it matters

OpenAlex reports 1 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

We investigate the influence of overconfidence and risk aversion on individual financial decision making in the experimental asset markets of the Smith, Suchanek and Williams (1988) type, with no informational asymmetries. Subjects, based on their pre-experimental overconfidence scores, were assigned to the two types of markets: least overconfident subjects formed five “rational” markets and most overconfident subjects formed five “overconfident” markets. The asset market experiment was followed by post hoc risk aversion measurement. Our results revealed that in the suggested setting, performance and trading activity were overconfidence dependent only for female participants. Mistakes in price forecasting, that are negatively correlated with overconfidence, could partially account for the increase in trading activity and losses. In the decreased sample differences in individual outcomes were overconfidence and not risk aversion driven.

Key concepts: Overconfidence effect, Risk aversion (psychology), Economics, Financial market, Asset (computer security), Behavioral economics, Financial economics, Loss aversion

Related papers

Back to paper searchBrowse research topicsOriginal source
Overconfidence, Risk Aversion and Individual Financial Decisions in Experimental Asset Markets — Research Paper | ScholarLens