Date of Graduation

8-2017

Document Type

Dissertation

Degree Name

Doctor of Philosophy in Economics (PhD)

Degree Level

Graduate

Department

Economics

Advisor/Mentor

Deck, Cary

Committee Member

Hao, Li

Second Committee Member

McGee, Peter J.

Third Committee Member

Jahedi, Salar

Keywords

Alcohol; Asset Markets; Insurance; Overconfidence; Overtrading

Abstract

This dissertation uses experimental evidence to explore the effects of overconfidence on economic decision making. In Chapter 1 I provide experimental evidence of the effects of alcohol on overconfidence and several other important tasks. I also explore the relationship between overconfidence and the behavior in the other tasks. The data from this experiment show that an alcohol level of 0.08 does not have a systemic effect on behavior and more importantly it does not affect one’s level of overconfidence. I also show that overconfidence is not significantly correlated with risk preferences, math, strategic behavior, anchoring, altruism, and food choices. In Chapter 2 I use feedback to establish a causal link between overconfidence and trading behavior. Feedback is used to eliminate the possibility for subjects to be overconfident about the accuracy of their signals. The data from this experiment show that overconfidence affects trading volume and profits, but when feedback is provided trading volume is no longer affected by overconfidence. This shows that there exists a causal relationship between overconfidence and trading volume. Lastly, Chapter 3 explores the role of overconfidence on insurance purchasing decisions. I show that overconfident people buy significantly less insurance. The stability of overconfidence using different measures and the relationship between overconfidence and risk is also explored. I find that different tasks do not elicit the same level of overconfidence and that risk preferences and overconfidence are not statistically significantly correlated.

Share

COinS