Are People Really Risk Seeking for Losses?

Kontek, Krzysztof (2009): Are People Really Risk Seeking for Losses?

Abstract

This short paper demonstrates that the claim of Cumulative Prospect Theory (CPT) that people are risk seeking for loss prospects, which confirmed a hypothetical assumption of the earlier Prospect Theory (PT), appears to be merely a result of using a specific form of the probability weighting function to estimate the power factor of the value function. Using experimental data and the form of the probability weighting function presented by CPT gives a power factor for losses of less than 1. This would mean that people are risk seeking for loss prospects. However, once more flexible, two-parameter forms are used, the power factor takes on values between 1.04 and 1.10. This, however, makes the value function convex, which indicates risk aversion. It follows that people are generally risk averse both for gains and for losses. This contradicts one of the main theses of Prospect Theory.

• Item Type: MPRA Paper
• Language: English
• Keywords: Prospect Theory; Value Function; Probability Weighting Function; Risk Attitude
• Subjects: D - Microeconomics > D0 - General > D03 - Behavioral Microeconomics: Underlying Principles; D - Microeconomics > D8 - Information, Knowledge, and Uncertainty > D81 - Criteria for Decision-Making under Risk and Uncertainty; C - Mathematical and Quantitative Methods > C9 - Design of Experiments > C91 - Laboratory, Individual Behavior; D - Microeconomics > D8 - Information, Knowledge, and Uncertainty > D87 - Neuroeconomics

References

Gonzales, R., Wu, G., (1999). On the Shape of the Probability Weighting Function. Cognitive Psychology, 38, pp 129-166.

Kahneman, D., Tversky, A., (1979). Prospect theory: An analysis of decisions under risk. Econometrica, 47, pp 313-327.

Prelec, D., (1998). The probability weighting function. Econometrica, 60, pp 497-529.

Tversky A., Kahneman D., (1992). Advances in Prospect Theory: Cumulative Representation of Uncertainty. Journal of Risk and Uncertainty, vol. 5(4), October, pp 297-323.