Asymptotic theory for maximum likelihood estimates in reduced-rank multivariate generalized linear models

Reduced-rank regression is a dimensionality reduction method with many applications. The asymptotic theory for reduced rank estimators of parameter matrices in multivariate linear models has been studied extensively. In contrast, few theoretical results are available for reduced-rank multivariate generalized linear models. We develop M-estimation theory for concave criterion functions that are maximized over parameter spaces that are neither convex nor closed. These results are used to derive the consistency and asymptotic distribution of maximum likelihood estimators in reduced-rank multivariate generalized linear models, when the response and predictor vectors have a joint distribution. We illustrate our results in a real data classification problem with binary covariates.     

Robust location estimation with missing data

In a missing data setting, we have a sample in which a vector of explanatory variables ${\bf x}_i$ is observed for every subject i, while scalar responses $y_i$ are missing by happenstance on some individuals. In this work we propose robust estimators of the distribution of the responses assuming missing at random (MAR) data, under a semiparametric regression model. Our approach allows the consistent estimation of any weakly continuous functional of the response's distribution. In particular, strongly consistent estimators of any continuous location functional, such as the median, L‐functionals and M‐functionals, are proposed. A robust fit for the regression model combined with the robust properties of the location functional gives rise to a robust recipe for estimating the location parameter. Robustness is quantified through the breakdown point of the proposed procedure. The asymptotic distribution of the location estimators is also derived. The proofs of the theorems are presented in Supplementary Material available online. The Canadian Journal of Statistics 41: 111–132; 2013 © 2012 Statistical Society of Canada     

High-risk sexual behavior among students of a minority-serving university in a community with a high HIV/AIDS prevalence

OBJECTIVE: The authors used a stratified cluster sampling design to inform campus sexually transmitted diseases prevention programs. PARTICIPANTS AND METHODS: They conducted a cross-sectional study of students (N = 1,130) at a large, urban, minority-serving university in South Florida using the 2004 National College Health Assessment Survey instrument. RESULTS: Of the 1,130 participants, 14.0% reported risky sexual behavior (having more than 1 sexual partner in 1 year and not using a condom the last time they had vaginal intercourse), and 11.9% reported consistent risky sexual behavior (having more than 1 partner in 1 year and not using condoms most or all of the time during the past 30 days). In multivariable analysis, past-month alcohol use had the strongest independent association with both risky and consistent risky sexual behavior. Illicit drug use was marginally significant in both models. CONCLUSIONS: Administrators should address the use of alcohol and other drugs on campus in planning interventions to promote safe sexual behavior.     

On the trail of line managers as talent spotters

The deliberate identification of talent is a critical process in most organizations. Performance ratings and senior management recommendations are its main sources, revealing the importance of line managers in the process. Nevertheless, some managers do have a talent for talent spotting and little is known about their behaviours as effective talent spotters. We explore, through social cognition, as part of the social psychology framework, the practices required to execute the talent identification process (TIP). An exploratory and qualitative method was adopted for this study; 21 line managers of an Argentine holding describe their experiences as talent spotters through semi-structured interviews. Findings show that the identification process is a complex and iterative one, in which several behaviours come into play, interrelated among each other. This interaction involves talent identification and talent development behaviours, deployed in a synergetic way. The relationship building behaviours set the foundations for the identification process to bear fruit. Distinct skills are necessary at different stages of the identification process for line managers to fulfil the talent spotter role. Finally, a model of the TIP is proposed.     

Revisiting clientelism: A network analysis of problem-solving networks in Argentina

Clientelism is a problem-solving network where brokers solve voter problems by providing material and non-material resources in exchange of political support.