On uniform approximation of Poisson variates by normal ones

In this paper the accuracy of the normal approximation to the Poisson is treated from the viewpoint of direct approximation of Poisson variables by normal ones. The conclusions that are derived on the accuracy of this approximation lead (among others) to very useful results on confidence limits for the mean of a linear combination of independent Poisson variables; these latter are employed in precise determination of the composition of a mixture of radioactive isotopes by means of a scintillation counter.     

Panel data regression for counts

A new panel data model for count data is introduced. We suggest alternative estimators, such as pseudo maximum likelihood and generalized method of moments, of structural and nuisance parameters. In addition, different test statistics of independence and overdispersion are obtained. The small sample performance of the estimators and tests are evaluated in Monte Carlo experiments. The model is applied to the number of days absent in Sweden 1981–1991 for a panel of Swedish male workers.     

The variability of rainfall acidity

Two years of rainfall acidity data for the eastern United States were analyzed. The data consist of rainfall-event pH measurements from a nine station monitoring network. A spatio-temporal stochastic model, including deterministic components for seasonal variation and rainfall washout, and stochastic components for spatial, temporal, and measurement variation, was fitted to the data. The fitted autocorrelation structure from this model was used, in the process known as Kriging, to obtain BLUE contour maps of seasonal and rainfall adjusted yearly average pH over the monitoring region.     

Interrelationships between fertility and marital dissolution: Results of a simultaneous logit model

Previous studies have failed to examine (a) the simultaneity of decisions to bear children and to dissolve a marriage and (b) the effects of the childbearing in the period just before dissolution on the likelihood of disruption. We attempted to resolve both problems by developing a simultaneous logit model of the interrelationship between the probability of separation and of having a birth during this period (when dissolution presumably is being considered). The model was estimated at successive durations of first marriage, using data for white women in the 1970 National Fertility Survey. The results indicate that childbearing patterns— number of children and age of youngest child at the beginning of the marital interval being studied and fertility during the interval—did not influence the likelihood of separation in simple or consistent ways over the marital life course, nor did marital strife (as indicated by separation) seem to affect childbearing throughout marriage.     

Scaling the prestige,authority, and income potential of college curricula

This paper develops the concept of “targeted education,” a theoretical ranking of college curricula, into a multidimensional framework. The new scales, based on the traditional stratification dimensions, prestige, authority, and income, are then used in a study of sex differences in the process of occupational achievement among men and women with college degrees. The targeted education scales predict occupational prestige and wages 7 years after the college degree, and they point out interesting differences between male and female attainment processes. In general, targeted education has a greater quantitative impact for men's occupational outcomes than for women's prestige and income, but results also suggest significant qualitative differences between men and women. A large proportion of women target their education toward, and end up in, an under-employed labor pool for the primary and secondary school system.     

The structure of consumer confidence in four EEC countries

This paper reports an exploration of the structure of consumer confidence in four of the countries participating in the EEC harmonized consumer attitudes surveys. The analysis is based on two EEC surveys in each of the countries concerned and uses the rotated solutions on the first three factors of the data. It was found that the different countries and surveys had similar factor structures which were interpreted as general economic conditions: personal financial circumstances; household durable buying intentions. The position of price expectations and attitudes to saving within these factor structures is also of interest in the interpretation of consumer behaviour.     

Influential observations in principal component analysis:a case study

A number of results have been derived recently concerning the influence of individual observations in a principal component analysis. Some of these results, particularly those based on the correlation matrix, are applied to data consisting of seven anatomical measurements on students. The data have a correlation structure which is fairly typical of many found in allometry. This case study shows that theoretical influence functions often provide good estimates of the actual changes observed when individual observations are deleted from a principal component analysis. Different observations may be influential for different aspects of the principal component analysis (coefficients, variances and scores of principal components); these differences, and the distinction between outlying and influential observations are discussed in the context of the case study. A number of other complications, such as switching and rotation of principal components when an observation is deleted, are also illustrated.     

Inequalities for linear combinations of gamma random variables

We study the behavior of the tail probabilities of weighted averages of certain independently and identically distributed random variables as the weights are varied. We show that the upper and lower tails are smallest when all the weights are equal. Our results apply to exponential, chi-squared, gamma, and Weibull random variables.