Diffusive nested sampling

We introduce a general Monte Carlo method based on Nested Sampling (NS), for sampling complex probability distributions and estimating the normalising constant. The method uses one or more particles, which explore a mixture of nested probability distributions, each successive distribution occupying ∼e ⁻¹ times the enclosed prior mass of the previous distribution. While NS technically requires independent generation of particles, Markov Chain Monte Carlo (MCMC) exploration fits naturally into this technique. We illustrate the new method on a test problem and find that it can achieve four times the accuracy of classic MCMC-based Nested Sampling, for the same computational effort; equivalent to a factor of 16 speedup. An additional benefit is that more samples and a more accurate evidence value can be obtained simply by continuing the run for longer, as in standard MCMC.     

Bayesian variable selection with spherically symmetric priors

We propose that Bayesian variable selection for linear parametrizations with Gaussian iid likelihoods should be based on the spherical symmetry of the diagonalized parameter space. Our r-prior results in closed forms for the evidence for four examples, including the hyper-g prior and the Zellner–Siow prior, which are shown to be special cases. Scenarios of a single-variable dispersion parameter and of fixed dispersion are studied, and asymptotic forms comparable to the traditional information criteria are derived. A simulation exercise shows that model comparison based on our r-prior gives good results comparable to or better than current model comparison schemes.     

Mean square error comparison among variance estimators with known coefficient of variation

The improved estimators for the population parameters were considered by several statisticians under various conditions. Recently Laheetharan and Wijekoon (Improved estimation of the population parameters when some additional information is available. Stat Papers doi:, 2008) demonstrated a generalized procedure for obtaining optimal shrunken estimators, and derived such estimators for both population mean and variance when coefficient of variation is known. In this article the mean square errors of those estimators were compared, and a numerical illustration was done using the scaled mean square error loss as used by Kanefuji and Iwase (Stat Papers 39:377–388, 1998) to understand the efficiency of the estimators with increasing sample size.     

Marital splits and income changes: Evidence from the British Household Panel Survey

We provide new evidence about what happens to people's incomes when their or their parents' marital union dissolves using longitudinal data from waves 1--4 of the British Household Panel Survey. Marital splits are accompanied by substantial declines in real income for separating wives and children on average, whereas separating husbands' real income on average changes much less. Results are shown to be robust to the choice of income de®nition and degree of economies of scale built into the household equivalence scale, and are validated with information about respondents' assessments of how their personal ®nancial circumstances changed. In addition we analyse the extent to which the welfare state mitigates the size of the income loss for women and children relative to men, and document the accompanying changes in social assistance bene®t receipt and paid work, and maintenance income receipt and payment.     

Value Accentuation or Convergence: Analysis of the Pattern and Process of College Student Change

Abstract

College student value change was analyzed for individual and group value accentuation or convergence among a longitudinal panel study of students in a traditional, large liberal arts college and a small, innovative living-learning subenvironment. Initially students in the innovative environment were more liberal, socally conscious, culturally sophisticated and intellectually oriented than their counterparts in the traditional setting, and, after two years, accentuation of these group differences had occurred. However, these group differences were not the result of accentuation of individual differences. Instead, increases in the student values were linked to interaction with peers and faculty on related topics and these patterns of interaction were more likely to occur in the living-learning subenvironment. Finally, those students who initially were lowest on a value and who frequently interacted with faculty or peers on related topics tended to experience the greatest value change.     

An examination of sexual strategies used by urban southern and rural Midwestern university women

The purposes of this study were (a) to determine whether women's use of persuasion, non‐physical coercion, and physical force strategies to obtain sex from a man varied between U.S. women from an urban Southern and rural Midwestern university and (b) to determine if sexual behavior history and early courtship behavior affected the use of these strategies (or not using any strategy). Women from the Midwest and the South did not differ in their use of sexual strategies. There were, however, other variables that accounted for differences in women's use of sexual strategies. Women who used persuasion strategies had fewer lifetime sexual partners than women who used any other strategy. Women who used physical force strategies reported a lower age at first intercourse and more early courtship behaviors than all others. Overall, the results indicated that sexual strategies are related to sexual behavior history and early courtship behaviors rather than cultural setting or demographics.     

Human Behavior and the Social Environment: Models,Metaphors and Maps for Applying Theoretical Perspectives to Practice

From Diagnosis to Treatment in Child Psychiatry. Group for the Advancement of Psychiatry. (New York: Jason Aronson, Inc., 1974.)

Psychotherapeutic Approaches to the Resistant Child. Dr. Richard Gardner. (New York: Jason Aronson, 1975.)     

Data-driven tests of uniformity on product manifolds

Data-driven versions of Sobolev tests of uniformity on compact Riemannian manifolds are reviewed and their large-sample asymptotic properties are given. A variant which is suitable for product manifolds is introduced. Data-driven goodness-of-fit tests of multivariate distributions are derived from data-driven tests of uniformity on tori.     

The Rate of Return on Investment in Wine

Is wine an investment alternative to stocks and bonds? First, we review prior studies regarding the pecuniary rate of return to wine and other alternatives to financial assets. Next, we calculate the rate of return to holding red Bordeaux wine from 1986 to 1996 using a repeat‐sale regression. Finally, we contrast the financial performance of wine, both on an aggregate basis and for various portfolios, to that of other asset classes. We conclude that wine does not yield greater returns than financial assets, especially when the volatility of returns and transaction costs are taken into account.