Sufficient Conditions for Expected Utility to Imply Mean-Standard Deviation Rankings: Empirical Evidence Concerning the Location and Scale Condition

• Published in: The Economic journal (London) Vol. 102; no. 410; pp. 91 - 106
• Main Authors: Meyer, Jack, Rasche, Robert H.
• Format: Journal Article
• Language: English
• Published: Cambridge Blackwell Publishers 01-01-1992; Cambridge University Press; Oxford University Press
• Subjects: Common stock; Distribution functions; Economic models; Economic statistics; Economic theory; Economics; Expected utility; Financial portfolios; Investment return rates; Portfolio management; Risk aversion; Sampling distributions; Security portfolios; Standard deviation; Statistical analysis; Utility theory
• ISSN: 0013-0133; 1468-0297
• DOI: 10.2307/2234854

A procedure is proposed for examining the location and scale (LS) condition. The Kolmogorov-Smirnov multi-sample test and observations from the random alternatives are used to determine whether or not the distribution functions describing the random alternatives are equal to one another except for location and scale. The test procedure is applied to the nonsystematic risk component of rate of return for portfolios of common stock. The test indicates that, for the portfolios and the stocks examined, the hypothesis that the expected utility (EU) efficient set of portfolios for risk averse investors is contained in the mean-standard (MS) efficient set cannot be rejected. When estimates of returns' distribution functions are treated as exact, MS and EU efficient sets do differ from one another, but not in a substantial way.