A Bayesian method for classification and discrimination

We discuss Bayesian analyses of traditional normal-mixture models for classification and discrimination. The development involves application of an iterative resampling approach to Monte Carlo inference, commonly called Gibbs sampling, and demonstrates routine application. We stress the benefits of exact analyses over traditional classification and discrimination techniques, including the ease with which such analyses may be performed in a quite general setting, with possibly several normal-mixture components having different covariance matrices, the computation of exact posterior classification probabilities for observed data and for future cases to be classified, and posterior distributions for these probabilities that allow for assessment of second-level uncertainties in classification.     

Efe fathers: one among many? A comparison of forager children's involvement with fathers and other males

The study reported describes Efe (pygmy) forager one-, two-, and three-year-olds' involvement with males. The Efe of northeastern Zaïre were chosen because their social organization allows us to examine hypotheses based on studies in Western, technologically complex societies about the distinctive role fathers play in the lives of their young children. Behavioral observations of Efe children's day-to-day activities with fathers, men and boys were recorded using a focal subject sampling technique (Altmann, 1974). Two behavioral measures were created to capture the extent to which males were involved with children: Social engagement describes males' involvement with children and social attention describes eavesdropping by children on males' everyday activities. Eight one-year-olds, 7 two-year-olds and 8 three-year-olds were each observed for six, one-hour observation sessions that were distributed evenly over the daylight hours. Data were analyzed using the traditional measure of involvement (e.g., adult males) and using a newly developed measure of the involvement of the average individual (e.g., average adult mate). Comparisons at each of the ages showed that fathers were consistently like other men in the extent to which children participated in social activities with them and watched their activities. Only fathers' level of social engagement declined significantly as children grew older. Boys' role relative to other males became increasingly distinctive as children aged. The findings suggest that Efe fathers may not be unique in the same sense assumed by Western study ideals, and raise questions about the special status given to fathers in Western theory and data. The patterning of mate involvement with children is discussed in terms of Efe community life, and in terms of Efe children's developing understanding of their relationship with fathers and other males.     

An introduction to econometric applications of empirical process theory for dependent random variables

This paper discusses some uses in econometrics of empirical process theory for dependent rendom variables. Examples considered include non-standard parametric hypotheses tests and semiparametric estimation. The application of bracketing functional limit results is discussed in some detail     

The historical decline of fertility in Eastern Europe

I test the hypothesis that the growth of autonomy was the factor responsible for the decline of marital fertility in eight Eastern European countries. By growth of autonomy I mean increased control over one's political, personal, religious, economic and reproductive life. This increased control was manifested in revolutions, democratic political reforms, nationalist movements and declining marital fertility. The political reforms were the result of the growth of autonomy but they also accelerated the dissemination of these new ideas among the populace contributing to further growth of autonomy. My hypothesis is generally supported by the historical data.     

Zusammenfassungen

Ohne Zusammenfassung German translations by Monika Lütke-Daldrup.     

A modified random-effect procedure for combining risk difference in sets of 2×2 tables from clinical trials

Summary Meta-analyses of sets of clinical trials often combine risk differences from several 2×2 tables according to a random-effects model. The DerSimonian-Laird random-effects procedure, widely used for estimating the populaton mean risk difference, weights the risk difference from each primary study inversely proportional to an estimate of its variance (the sum of the between-study variance and the conditional within-study variance). Because those weights are not independent of the risk differences, however, the procedure sometimes exhibits bias and unnatural behavior. The present paper proposes a modified weighting scheme that uses the unconditional within-study variance to avoid this source of bias. The modified procedure has variance closer to that available from weighting by ideal weights when such weights are known. We studied the modified procedure in extensive simulation experiments using situations whose parameters resemble those of actual studies in medical research. For comparison we also included two unbiased procedures, the unweighted mean and a sample-size-weighted mean; their relative variability depends on the extent of heterogeneity among the primary studies. An example illustrates the application of the procedures to actual data and the differences among the results. This research was supported by Grant HS 05936 from the Agency for Health Care Policy and Research to Harvard University.