Are Risk-Attitudes Related Across Domains and Response Modes?
This paper examines experimentally the nature of people's risk-attitudes across different payoff domains and response modes. Only simple gambles are examined, entailing just two monetary outcomes. The main issue of interest is to what extent risk-preferences in one domain or response mode predict anything (beyond chance) about risk-preferences in another domain or mode. Three domains are examined: gain, mixed and loss. The three response modes used are certainty equivalence (CE), probability equivalence (PE) and outcome equivalence (OE) judgments. In general, weak associations were found among ordinal risk-preferences within-subjects across domains, especially with respect to losses. To make the parametric responses (i.e., the CE, PE and OE judgments) comparable across domains and subjects, linear as well as utility-based risk-measures were examined. In the gain and loss domains, the linear risk-premia measures exhibited higher CE-PE correlations (within domain) than the utility-based measures. Using a multitrait-multimethod comparison, the highest correlations were found within domains across response modes. The main findings are (1) strong risk-aversion for gain and mixed gambles, (2) risk-seeking for symmetric loss gambles, although less pronounced, (3) low correlation of risk-preferences within subjects across domains, (4) high convergent validity of response methods within domain, and (5) increased risk-aversion for loss (but not gain) gambles when using real payoffs.
Volume (Year): 36 (1990)
Issue (Month): 12 (December)
|Contact details of provider:|| Postal: 7240 Parkway Drive, Suite 300, Hanover, MD 21076 USA|
Web page: http://www.informs.org/
More information through EDIRC
When requesting a correction, please mention this item's handle: RePEc:inm:ormnsc:v:36:y:1990:i:12:p:1451-1463. See general information about how to correct material in RePEc.
For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: (Mirko Janc)
If references are entirely missing, you can add them using this form.