Users (or would-be users) of exponential expected utility often seek a concrete, intuitive meaning for the risk tolerance coefficient (RT) that they can grasp and explain to others easily. This paper shows an interpretation of RT as the maximum loss the decision maker is willing to be exposed to at a stated probability level, regardless of the upside potential. As an example, if you are facing a portfolio of projects having a 1 in 20 chance of total loss and you indicate that L is the maximum loss you would tolerate, then your risk tolerance is L/3. Other such examples, which may be better suited to other situations, are presented. In some contexts, this interpretation may be congruent with the way individuals naturally think about their risk-taking propensity, for example, as willingness to be exposed to losses or as in value-at-risk. This interpretation can also be helpful in determining whether assuming exponential utility is adequate for the situation being analyzed, and in eliciting the value of RT. The merits of this approach for thinking about RT are discussed.
[1]
R. Howard.
Proximal Decision Analysis
,
1971
.
[2]
M. Machina.
"Expected Utility" Analysis without the Independence Axiom
,
1982
.
[3]
Ronald A. Howard,et al.
Decision analysis: practice and promise
,
1988
.
[4]
J. Pratt.
RISK AVERSION IN THE SMALL AND IN THE LARGE11This research was supported by the National Science Foundation (grant NSF-G24035). Reproduction in whole or in part is permitted for any purpose of the United States Government.
,
1964
.
[5]
K. Arrow.
Essays in the theory of risk-bearing
,
1958
.
[6]
D. Duffie,et al.
An Overview of Value at Risk
,
1997
.
[7]
C. Kirkwood.
Approximating Risk Aversion in Decision Analysis Applications
,
2004
.
[8]
James E. Smith,et al.
Risk Sharing, Fiduciary Duty, and Corporate Risk Attitudes
,
2004,
Decis. Anal..