Robust estimation of variance components

New robust estimates for variance components are introduced. Two simple models are considered: the balanced one-way classification model with a random factor and the balanced mixed model with one random factor and one fixed factor. However, the method of estimation proposed can be extended to more complex models. The new method of estimation we propose is based on the relationship between the variance components and the coefficients of the least-mean-squared-error predictor between two observations of the same group. This relationship enables us to transform the problem of estimating the variance components into the problem of estimating the coefficients of a simple linear regression model. The variance-component estimators derived from the least-squares regression estimates are shown to coincide with the maximum-likelihood estimates. Robust estimates of the variance components can be obtained by replacing the least-squares estimates by robust regression estimates. In particular, a Monte Carlo study shows that for outlier-contaminated normal samples, the estimates of variance components derived from GM regression estimates and the derived test outperform other robust procedures.     

Inferring large graphs using $$\ell _1$$-penalized likelihood

We address the issue of recovering the structure of large sparse directed acyclic graphs from noisy observations of the system. We propose a novel procedure based on a specific formulation of the \(\ell _1\)-norm regularized maximum likelihood, which decomposes the graph estimation into two optimization sub-problems: topological structure and node order learning. We provide convergence inequalities for the graph estimator, as well as an algorithm to solve the induced optimization problem, in the form of a convex program embedded in a genetic algorithm. We apply our method to various data sets (including data from the DREAM4 challenge) and show that it compares favorably to state-of-the-art methods. This algorithm is available on CRAN as the R package GADAG.     

Variational Bayes with synthetic likelihood

Synthetic likelihood is an attractive approach to likelihood-free inference when an approximately Gaussian summary statistic for the data, informative for inference about the parameters, is available. The synthetic likelihood method derives an approximate likelihood function from a plug-in normal density estimate for the summary statistic, with plug-in mean and covariance matrix obtained by Monte Carlo simulation from the model. In this article, we develop alternatives to Markov chain Monte Carlo implementations of Bayesian synthetic likelihoods with reduced computational overheads. Our approach uses stochastic gradient variational inference methods for posterior approximation in the synthetic likelihood context, employing unbiased estimates of the log likelihood. We compare the new method with a related likelihood-free variational inference technique in the literature, while at the same time improving the implementation of that approach in a number of ways. These new algorithms are feasible to implement in situations which are challenging for conventional approximate Bayesian computation methods, in terms of the dimensionality of the parameter and summary statistic.     

Men’s Migration and Women’s Fertility in Rural Mozambique

Labor migration profoundly affects households throughout rural Africa. This study looks at how men’s labor migration influences marital fertility in a context where such migration has been massive while its economic returns are increasingly uncertain. Using data from a survey of married women in southern Mozambique, we start with an event-history analysis of birth rates among women married to migrants and those married to nonmigrants. The model detects a lower birth rate among migrants’ wives, which tends to be partially compensated for by an increased birth rate upon cessation of migration. An analysis of women’s lifetime fertility shows that it decreases as the time spent in migration by their husbands accrues. When we compare reproductive intentions stated by respondents with migrant and nonmigrant husbands, we find that migrants’ wives are more likely to want another child regardless of the number of living children, but the difference is significant only for women who see migration as economically benefiting their households. Yet, such women are also significantly more likely to use modern contraception than other women. We interpret these results in light of the debate on enhancing versus disrupting effects of labor migration on families and households in contemporary developing settings.     

Attitude and Propensity to Divorce in Iran: Structural and Ideational Determinants

Divorce has been increasing worldwide, even in societies where religious impediments to it are strongest. In the Islamic Republic of Iran, divorce rates have risen rapidly since the mid-1990s. In this article, we investigate the attitude and propensity of young Iranians toward divorce and relate these attitudes and propensities to structural and ideational factors. The data are drawn from a survey of 720 married people ages 15 to 29 conducted in the city of Tehran in 2014. The results show that almost half of respondents approved of divorce as a solution for marital problems and one fifth of them have high propensity to divorce. Multivariate analyses indicate that approval of and high propensity toward divorce are significantly associated with ideational factors—namely individualism, secularism, and gender egalitarian views—and structural factors—including education, wife’s employment, and household economic insecurity—even after controlling for demographic variables. We discuss the implications of these findings for the understanding of marital stability in this rapidly changing context.     

From partisan war to the ethnography of European Turkey: the Balkan career of Ivan Liprandi, 1790–1880

The Balkan entanglements of the Great Powers have long interested historians of war, diplomacy, and nation-building in South-Eastern Europe. Although tsarist officers played a central role in Russian policies in the region, historians have rarely treated their writings as expressions of specifically military concerns and preoccupations. The present article seeks to fill this gap by reconstructing the Balkan career of Ivan Liprandi, a Russian officer, partisan leader, and self-styled expert on the European part of the Ottoman Empire during the middle decades of the nineteenth century. The article traces Liprandi’s effort to turn his direct experience of partisan warfare into knowledge and place that knowledge at the disposal of the Russian military command. Liprandi’s remarkable Balkan career testifies to the growing interest of the Imperial Russian military in the ethno-confessional profile and political attitudes of the local population as factors contributing to victory or defeat. Liprandi’s statistical and ethnographic writings on the Balkans are also indicative of the nexus between the military and civilian forms of knowledge that emerged in the middle decades of the nineteenth century.     

Video Lottery is the Most Harmful Form of Gambling in Canada

This paper summarizes the degree to which different forms of legal gambling contribute to Problem and Pathological Gambling (PPG) in Canada. Legal gambling activities were compared using meta-analysis of publicly available data concerning Canada’s legal gambling industry. The majority of revenues in the decade spanning 2002–2012 were drawn from Video Lottery Terminals and casino slot machines. Population surveys indicated that three quarters of Canadians reported some form of past-year gambling participation, but most did not play Electronic Gambling Machines. Annual revenues divided by estimated numbers of participants in various gambling activities showed that Video Lottery players spent more money on average than did participants in other forms of gambling. The relative risk of PPG was higher among Video Lottery players than it was for other common forms of gambling. Results from a community study of frequent Video Lottery players showed that the risk of frequent players reporting symptoms of PPG was elevated if they reported playing weekly, spending $50 or more per session, or playing for more than an hour per session. These studies provide converging evidence that Video Lottery is more hazardous to consumers than other forms of gambling that are commonly practised in Canada.     

Determined to stop? Longitudinal analysis of the desire to have no more children in rural Mozambique

Classic demographic theories conceptualize desired family size as a fixed goal that guides fertility intentions over the childbearing years. However, a growing body of research shows that fertility plans, even nominally long-term plans for completed childbearing, change in response to short-term conditions. Because of data limitations, much of this research has focused on low-fertility contexts, but short-term conditions are likely to be even more important in high-fertility contexts. This paper uses three waves of survey data collected in rural Mozambique to study predictors of the desire to stop childbearing in a context of relatively high fertility and high individual and social instability. We use fixed effects models to assess how women’s desires to stop childbearing are shaped by demographic factors, household economic conditions, and health status, controlling for constant individual characteristics. Results provide evidence that fertility desires both reflect stable underlying goals and evolve in response to individual and social circumstances.