Outlier detection in contingency tables using decomposable graphical models

For high-dimensional data, it is a tedious task to determine anomalies such as outliers. We present a novel outlier detection method for high-dimensional contingency tables. We use the class of decomposable graphical models to model the relationship among the variables of interest, which can be depicted by an undirected graph called the interaction graph. Given an interaction graph, we derive a closed-form expression of the likelihood ratio test (LRT) statistic and an exact distribution for efficient simulation of the test statistic. An observation is declared an outlier if it deviates significantly from the approximated distribution of the test statistic under the null hypothesis. We demonstrate the use of the LRT outlier detection framework on genetic data modeled by Chow–Liu trees.     

Sensitivity analysis of partially linear models with response missing at random

This article investigates case-deletion influence analysis via Cook’s distance and local influence analysis via conformal normal curvature for partially linear models with response missing at random. Local influence approach is developed to assess the sensitivity of parameter and nonparametric estimators to various perturbations such as case-weight, response variable, explanatory variable, and parameter perturbations on the basis of semiparametric estimating equations, which are constructed using the inverse probability weighted approach, rather than likelihood function. Residual and generalized leverage are also defined. Simulation studies and a dataset taken from the AIDS Clinical Trials are used to illustrate the proposed methods.     

Leaving Science: Gender Differences in Career Continuity Following a Job Loss

This study investigates gender differences in the postdisplacement experience of nonacademic science and engineering (S&E) workers. Using a pooled sample created from the Displaced Worker Surveys conducted between 1994 and 2008, it finds that (1) this S&E work force is particularly vulnerable to job loss and potential career disruption; (2) displaced female S&E workers are more likely than comparable male workers to exit the work force, a gender difference that is conditional on and explained by marital and parental status; and (3) reemployed female S&E workers are also more likely to leave science for non‐S&E occupations, but this gender difference is limited to unmarried workers. A concluding section discusses the implications of these findings for interpreting gender differences in career outcomes.     

LIMIT THEOREMS FOR SUBCRITICAL AGE-DEPENDENT BRANCHING PROCESSES WITH TWO TYPES OF IMMIGRATION

ABSTRACT

For the classical subcritical age-dependent branching process the effect of the following two-type immigration pattern is studied. At a sequence of renewal epochs a random number of immigrants enters the population. Each subpopulation stemming from one of these immigrants or one of the ancestors is revived by new immigrants and their offspring whenever it dies out, possibly after an additional delay period. All individuals have the same lifetime distribution and produce offspring according to the same reproduction law. This is the Bellman-Harris process with immigration at zero and immigration of renewal type (BHPIOR). We prove a strong law of large numbers and a central limit theorem for such processes. Similar conclusions are obtained for their discrete-time counterparts (lifetime per individual equals one), called Galton-Watson processes with immigration at zero and immigration of renewal type (GWPIOR). Our approach is based on the theory of regenerative processes, renewal theory and occupation measures and is quite different from those in earlier related work using analytic tools.     

On some probability distributions for forward birth interval

Parents’ decisions to have children are modeled by a simple stopping rule that describes the probability of having another child as a function of the number of boys and girls already born to the parents. Because the stopping rule depends on the sex of the offspring, the rule may introduce a correlation between sex of offspring and the number of siblings the offspring has. When this is coupled with a correlation between number of siblings and well‐being, a correlation between sex and well‐being may emerge despite equal treatment of the two sexes within each family. The author provides sufficient conditions on a stopping rule for it to be sexist in the sense that the average well‐being of one sex is higher than that of the other sex.     

Statistical modeling of selected aspects of the childbearing process with application to world fertility survey countries∗

The childbearing process should be monitored in developing countries experiencing high population growth rates and high levels of maternal and infant mortality. A mathematical model for estimation of certain aspects of the childbearing process, which requires only data on age‐specific fertility rates, is developed. Synthetic maternal childbearing indices, namely, mean ages at first and last birth, length of reproductive life span, inter‐birth spacing, and proportion of childless women, in addition to the well‐known mean age at childbearing, for the WFS countries are obtained using the proposed model. The indices are free from age truncation effects, and, under certain assumptions, provide information about a cohort's completed fertility before the women stop reproducing. The effects of women's residence and education on fertility are also examined.     

Goodness of fit tests with misclassified data

The most popular goodness of fit test for a multinomial distribution is the chi-square test. But this test is generally biased if observations are subject to misclassification, In this paper we shall discuss how to define a new test procedure when we have double sample data obtained from the true and fallible devices. An adjusted chi-square test based on the imputation method and the likelihood ratio test are considered, Asymptotically, these two procedures are equivalent. However, an example and simulation results show that the former procedure is not only computationally simpler but also more powerful under finite sample situations.     

On some tests in manocova model with different dispersion matrices

In this article we have considered the problem of testing linear hypothesis in MANOCOVA model with different dispersion Matrices by two test procedures in the line of Anderson (1963) and fihargava CI971). Also efficiencies of the two tests have often compared.     

Thank you!

The study focuses on the question of how HRD personnel employed in a large multinational company perceive learning at work. We are interested in how HR development staff describe learning at the individual, collective and organizational levels. The participants were Finnish and Chinese human resource professionals (n?=?17) who used an asynchronous web-based tool to study adult education. The empirical data consist of all the texts sent to a web-based discussion forum. The findings showed that the HRD practitioners' views of learning covered quite evenly all three levels of organizational learning. The practitioners paid special attention to practical learning taking place in connection with everyday work activities, to communication and collaborative learning and to the development of a learning organization. This reflects the multifaceted nature of learning at work and the necessity of using HRD to integrate individual- and organizational-level needs in order to enhance organizational learning. The participants frequently reported that they found this a demanding task that required a clear 'understanding of humans, people in the organizations and their learning'. If organizational learning is to meet its current challenges it should take note of messages like these from HRD practitioners, which emphasise the diversity of learning experiences. Promoting a shared understanding and especially awareness and recognition of the fundamental issues associated with learning at work is a possible first step.     

An Integrated Experiential Approach to Treating Young People Who Sexually Abuse

SUMMARY

This article promotes the use of an integrated (holistic) approach to treating juvenile sexual offenders. An integrated model takes into account the fact that: (a) youth are resilient, (b) youth progress through various stages of development, (c) these stages are often arrested as a result of trauma, child abuse and neglect, and attachment disorders, (d) humanistic approaches and the therapeutic relationship are essential to the healing and recovery process, (e) youth learn and work with a variety of learning styles and multiple intelligences, (f) many traditional assessment and treatment approaches can be modified and blended with an integrated approach, and (g) the use of experiential treatments can have a positive and profound impact in treating youth with sexual behavior problems.