A skew logistic distribution as an alternative to the model of van Staden and King

The logistic distribution is a simple distribution possessing many useful properties and has been used extensively for analyzing growth. Recently, van Staden and King proposed a quantile-based skew logistic distribution. In this paper, we introduce an alternative skew logistic distribution. We then establish recurrence relations for the computation of the single and product moments of order statistics from the standard skew logistic distribution by using the moments of order statistics from the standard half logistic distribution. These enable an efficient computation of means, variances and covariances of order statistics from the skew logistic distibution for all sample sizes. The results become useful in determining the best linear unbiased estimators of the location and scale paramters of the skew logistic distribution. Finally, we provide an example to illustrate the usefulness of the developed model and then compare its fit with that provided by the model of van Staden and King.     

Order statistics and their concomitants from multivariate normal mean–variance mixture distributions with application to Swiss Markets Data

In this article, by considering a multivariate normal mean–variance mixture distribution, we derive the exact joint distribution of linear combinations of order statistics and their concomitants. From this general result, we then deduce the exact marginal and conditional distributions of order statistics and their concomitants arising from this distribution. We finally illustrate the usefulness of these results by using a Swiss markets dataset.     

Likelihood inference for Type I bivariate Pólya–Aeppli distribution

This article discusses likelihood inference for the Type I bivariate Pólya–Aeppli distribution. The Type I bivariate Pólya–Aeppli distribution was derived by Minkova and Balakrishnan by using compounding with geometric random variables and the trivariate reduction method. They also discussed the moment estimation of the parameters of the Type I bivariate Pólya–Aeppli distribution. Here, we carry out a simulation study to compare the performance of the developed Maximum Likelihood Estimation (MLE) method with the moment estimation. The obtained results show that, through the MLEs require more computational time compared to the moment estimates (MoM), the MLEs perform better, in most of the settings, than the MoM. Finally, we apply the Type I bivariate Pólya–Aeppli model to a real dataset containing the frequencies of railway accidents in two subsequent six-year periods for the purpose of illustration. We also carry out some hypothesis tests using the Wald test statistic. From these results, we conclude that the two variables belong to the same univariate Pólya–Aeppli distribution, but are correlated.     

Multivariate Bayesian U-type asymmetric designs for non parametric response surface prediction under correlated errors

The objective of this paper is to study U-type designs for Bayesian non parametric response surface prediction under correlated errors. The asymptotic Bayes criterion is developed in terms of the asymptotic approach of Mitchell et al. (1994 Mitchell, T., Sacks, J., Ylvisaker, D. (1994). Asymptotic Bayes criteria for nonparametric response surface design. Ann. Stat. 22:634–651.[Crossref], [Web of Science ®] , [Google Scholar]) for a more general covariance kernel proposed by Chatterjee and Qin (2011 Chatterjee, K., Qin, H. (2011). Generalized discrete discrepancy and its applications in experimental designs. J. Stat. Plann. Inference 141:951–960.[Crossref], [Web of Science ®] , [Google Scholar]). A relationship between the asymptotic Bayes criterion and other criteria, such as orthogonality and aberration, is then developed. A lower bound for the criterion is also obtained, and numerical results show that this lower bound is tight. The established results generalize those of Yue et al. (2011 Yue, R.X., Qin, H., Chatterjee, K. (2011). Optimal U-type design for Bayesian nonparametric multiresponse prediction. J. Stat. Plann. Inference 141:2472–2479.[Crossref], [Web of Science ®] , [Google Scholar]) from symmetrical case to asymmetrical U-type designs.     

Ethnic dimensions of gender differentials in mortality in Malaysia

Higher mortality rates among males are a common occurrence across different cultures and countries. The causes of this higher mortality can be biological as well as behavioural in nature. The biological evidence applies across all nations and communities, but the behavioural causes, arising from the decision processes and communication strategies of individuals, will necessarily have cultural and environmental dimensions that change with time. This study examines gender disparities in mortality across ethnicity and time in Malaysia. The study shows that there is a consistent gender differential across time but it has widened for the Malays and the Indians and narrowed for the Chinese. Most importantly, it has widened considerably for young adults. Analysis of the leading causes of death show that young adult males are more likely to engage in risk-taking behaviour, and that the related causes and the extent of such causes vary across the ethnic groups.     

Hand infection associated with clenched fist syndrome in residents of long-term care facilities

1. Clenched fist syndrome is a clinical entity in which no organic disease can be found. 2. The syndrome usually follows a minor inciting incident and is associated with swelling, pain, and paradoxical stiffness. 3. Patients with severe forms of clenched fist syndrome may experience infections of the palm or joints.     

Some simple nonparametric methods to test for perfect ranking in ranked set sampling

A lot of research on ranked set sampling (RSS) is based on the assumption that the ranking is perfect. Hence, it is necessary to develop some tests that could be used to validate this assumption of perfect ranking. In this paper, we introduce some simple nonparametric methods for this purpose. We specifically define three test statistics, _Nk,_Sk$N_k,S_k$ and _Ak$A_k$, based on one-cycle RSS, which are all associated with the ordered ranked set sample (ORSS). We then derive the exact null distributions and exact power functions of all these tests. Next, by using the sum or the maximum of each statistic over all cycles, we propose six test statistics for the case of multi-cycle RSS. We compare the performance of all these tests with that of the Kolmogorov–Smirnov test statistic proposed earlier by Stokes and Sager [1988. Characterization of a ranked-set sample with application to estimating distribution functions. J. Amer. Statist. Assoc. 83, 35–42] and display that all proposed test statistics are more powerful. Finally, we present an example to illustrate the test procedures discussed here.     

Is the Burgess concentric zonal theory of spatial differentiation still applicable to urban Canada?

L'auteur examine les données par centre de recensement provenant des recensements de 1961,1971 et 1981 dans le but d'étudier, dans les quatorze agglomérations les plus importantes au Canada, la persistance de la répartition de la population en zones concentriques selon la situation socio-économique et la taille de la famille prédite par la théorie de Burgess. II conclut que même si le caractère des zones n'est plus tout à fait le même que dans le Canada d'antan, le phénomène de la stratification progressive a non seulement persisté, mais s'est peut-être même accentué en ce qui a trait à la situation socio-économique et à la taille des families. Cette persistance s'explique, d'une part, par les valeurs sociales, qui continuent à attacher beaucoup d'importance à la possession d'une maison unifamiliale et, d'autre part, par l'aménagement et le maintien d'un réseau routier qui permet aux gens de travailler dans les grandes villes canadiennes tout en vivant en banlieue ou à la campagne.
Using census tract data from the censuses of 1961,1971 and 1981, this paper examines the durability of concentric zonal patterns in socio-economic status and family size predicted by Burgess' theory in the 14 largest metropolitan areas of Canada. Although the character of the zones may have changed over time in modern day Canada, it was found that the gradient pattern itself not only persists but may have intensified as far as socio-economic status and family size are concerned. This persistence is attributed to the strong societal norm of owning a single family detached dwelling and the development and maintenance of roadways which still make it worthwhile to commute to work in Canadian cities.     

Educational attainment, age at first birth and lifetime fertility: an analysis of Canadian fertility survey data

Dans cet article, l'auteur réexamine de manière détaillée les rapports entre le niveau d'instruction, lâge à la naissance du premier-né et la descendance finale des femmes au Canada dans les années 1980, compte tenu des changements en matière d'options socio-professionnelles pour les femmes qui se sont produits au cours de la dernière génération. Les données employées sont tirées de l'Enquête nationale sur la fécondité au Canada de 1984 et concernent les femmes âgées de 35 à 49 ans (n= 2083). Les résultats révèlent une corrélation plus forte entre le niveau d'instruction des femmes et le choix du moment de la première grossesse qu'entre le niveau d'instruction et la descendance finale. De plus, cette influence du niveau d'instruction sur le choix du moment de la première grossesse semble être l'un des grands déterminants de la tendance générale. Cependant, l'étude a permis de dégager une nouvelle tendance démographique importante: même si elles ont moins de 20 ans quand elles donnent naissance à leur premier enfant, les femmes de la plus jeune cohorte (âgées de 35 à 39 ans) ont eu moins de trois enfants au total. Tout laisse prévoir le maintien chez les femmes canadiennes des tendances récentes au mariage tardif, à une progéniture moins nombreuse, à des niveaux d'instruction plus élevés et à une participation accrue à la population active. The purpose of this paper is to re-examine in some detail the relationship between education, age at first birth and life-time fertility of women in Canada in the 1980s. This re-examination is related to the changes that have taken place in the past generation in life course paths for women. The data in this paper are developed from the Canadian National Fertility Survey of 1984, focussing on women 35–49 years of age (N= 2,083). The findings show that it is the timing of first birth rather than the number of children ever born that is more related to educational attainment of women. In addition, it appears to be the effect of education on first birth timing rather than lifetime fertility that is of key importance in the overall pattern. However, an important new demographic trend is observable. Even when there is an early first birth (under the age of 20), the youngest cohort of women (age 35–39) had fewer than three children over the reproductive lifetime. It is likely that the recent trends of later marriage, smaller numbers of children, later age at first birth, and increasing levels of education and labour force activity will continue for women in Canada.     

On the maximum likelihood estimation of the location and scale parameters of exponential distribution based on multiply Type II censored samples

The maximum likelihood (ML) estimation of the location and scale parameters of an exponential distribution based on singly and doubly censored samples is given. When the sample is multiply censored (some middle observations being censored), however, the ML method does not admit explicit solutions. In this case we present a simple approximation to the likelihood equation and derive explicit estimators which are linear functions of order statistics. Finally, we present some examples to illustrate this method of estimation.