Ambiguity framed

In his exposition of subjective expected utility theory, Savage (1954) proposed that the Allais paradox could be reduced if it were recast into a format which made the appeal of the independence axiom of expected utility theory more transparent. Recent studies consistently find support for this prediction. We consider a salience-based choice model which explains this frame-dependence of the Allais paradox. We then derive the novel prediction that the presentation format responsible for reductions in Allais-style violations of expected utility theory will also reduce Ellsberg-style violations of subjective expected utility theory. This format makes the appeal of Savage’s “sure thing principle” more transparent. We design an experiment to test this prediction and find strong support for such frame-dependence of ambiguity aversion in Ellsberg-style choices. In particular, we observe markedly less ambiguity-averse behavior in Savage’s matrix format than in a more standard “prospect” format. This finding poses a new challenge for the leading models of ambiguity aversion.     

Inferences Based on a Skipped Correlation Coefficient

The most popular method for trying to detect an association between two random variables is to test H ₀ ?:?ρ=0, the hypothesis that Pearson's correlation is equal to zero. It is well known, however, that Pearson's correlation is not robust, roughly meaning that small changes in any distribution, including any bivariate normal distribution as a special case, can alter its value. Moreover, the usual estimate of ρ, r, is sensitive to only a few outliers which can mask a true association. A simple alternative to testing H ₀ ?:?ρ =0 is to switch to a measure of association that guards against outliers among the marginal distributions such as Kendall's tau, Spearman's rho, a Winsorized correlation, or a so-called percentage bend correlation. But it is known that these methods fail to take into account the overall structure of the data. Many measures of association that do take into account the overall structure of the data have been proposed, but it seems that nothing is known about how they might be used to detect dependence. One such measure of association is selected, which is designed so that under bivariate normality, its estimator gives a reasonably accurate estimate of ρ. Then methods for testing the hypothesis of a zero correlation are studied.     

A random-effects model for cycle viability in fertility studies

The authors describe a random-effects fertility model based upon the assumption that the menstrual cycle viability probability varies from couple to couple according to a beta distribution. An EM algorithm is used to fit the model. The proposed estimating procedure is fully expandable to allow covariate effects on the beta variate. The method can be applied generally whenever dependency among Bernoulli trials is induced by a susceptibility state and the outcomes can be observed only in the aggregate. Based upon data from a cohort of 221 couples with no known fertility problems who were attempting pregnancy, cycle viability was found to be heterogeneous among couples. Stratification on the presence or absence of prenatal exposure of the woman to her mother's cigarette smoking revealed a statistically significant difference in the two-cycle viability distributions. Differences are discussed in the interpretation of the beta model compared to the marginal approach based upon generalized estimating equations.     

Confidence interval estimation for the population coefficient of variation using ranked set sampling: a simulation study

In this paper, an evaluation of the performance of several confidence interval estimators of the population coefficient of variation (τ) using ranked set sampling compared to simple random sampling is performed. Two performance measures are used to assess the confidence intervals for τ, namely: width and coverage probabilities. Simulated data were generated from normal, log-normal, skew normal, Gamma, and Weibull distributions with specified population parameters so that the same values of τ are obtained for each distribution, with sample sizes n=15, 20, 25, 50, 100. A real data example representing birth weight of 189 newborns is used for illustration and performance comparison.     

Hypothesis Testing in Multivariate Linear Models with Randomly Missing Data

Tests for mean equality proposed by Weerahandi (1995) and Chen and Chen (1998), tests that do not require equality of population variances, were examined when data were not only heterogeneous but, as well, nonnormal in unbalanced completely randomized designs. Furthermore, these tests were compared to a test examined by Lix and Keselman (1998), a test that uses a heteroscedastic statistic (i.e., Welch, 1951) with robust estimators (20% trimmed means and Winsorized variances). Our findings confirmed previously published data that the tests are indeed robust to variance heterogeneity when the data are obtained from normal populations. However, the Weerahandi (1995) and Chen and Chen (1998) tests were not found to be robust when data were obtained from nonnormal populations. Indeed, rates of Type I error were typically in excess of 10% and, at times, exceeded 50%. On the other hand, the statistic presented by Lix and Keselman (1998) was generally robust to variance heterogeneity and nonnormality.     

Global comparisons of medians and other quantiles in a one-way design when there are tied values

For J ? 2 independent groups, the article deals with testing the global hypothesis that all J groups have a common population median or identical quantiles, with an emphasis on the quartiles. Classic rank-based methods are sometimes suggested for comparing medians, but it is well known that under general conditions they do not adequately address this goal. Extant methods based on the usual sample median are unsatisfactory when there are tied values except for the special case J = 2. A variation of the percentile bootstrap used in conjunction with the Harrell–Davis quantile estimator performs well in simulations. The method is illustrated with data from the Well Elderly 2 study.     

New Results on the Small-Sample Properties of Some Robust Univariate Estimators of Location

This article compares eight estimators in terms of relative efficiencies with the univariate mean, some of which have not been compared previously. Four estimators, when testing hypotheses, are compared in terms of actual Type I errors. In terms of point estimation, the modified one-step M-estimator, one-step M-estimator, and rfch estimator are found to be the three best choices depending on the proportion of outliers. In terms of actual Type I errors, the modified one-step M estimator's and rfch estimator's level was between.045 and.055 in 5 out of 7 situations when real data were used in simulations.     

A comment on approximating the x2 distribution in the equiprobable case

In an article appearing in this journal, Smith, et al. (1979) reported that a beta approximation of the X² distribution in the equiprobable case did not perform well. This brief note points out that they used the wrong moments and range in their approximation. It is suggested that when this problem is cor-rected, a beta approximation performs better than the chi-square or adjusted chi-square when k ≥ 3 and n is not too small.