Marginal zero-inflated regression models for count data

Data sets with excess zeroes are frequently analyzed in many disciplines. A common framework used to analyze such data is the zero-inflated (ZI) regression model. It mixes a degenerate distribution with point mass at zero with a non-degenerate distribution. The estimates from ZI models quantify the effects of covariates on the means of latent random variables, which are often not the quantities of primary interest. Recently, marginal zero-inflated Poisson (MZIP; Long et al. [A marginalized zero-inflated Poisson regression model with overall exposure effects. Stat. Med. 33 (2014), pp. 5151–5165]) and negative binomial (MZINB; Preisser et al., 2016) models have been introduced that model the mean response directly. These models yield covariate effects that have simple interpretations that are, for many applications, more appealing than those available from ZI regression. This paper outlines a general framework for marginal zero-inflated models where the latent distribution is a member of the exponential dispersion family, focusing on common distributions for count data. In particular, our discussion includes the marginal zero-inflated binomial (MZIB) model, which has not been discussed previously. The details of maximum likelihood estimation via the EM algorithm are presented and the properties of the estimators as well as Wald and likelihood ratio-based inference are examined via simulation. Two examples presented illustrate the advantages of MZIP, MZINB, and MZIB models for practical data analysis.     

Tail asymptotic of discounted aggregate claims with compound dependence under risky investment

This paper considers the tail asymptotic of discounted aggregate claims with compound dependence under risky investment. The price of risky investment is modeled by a geometric Lévy process, while claims are modeled by a one-sided linear process whose innovations further obeying a so-called upper tail asymptotic independence. When the innovations are heavy tailed, we derive some uniform asymptotic formulas. The results show that the linear dependence has significant impact on the tail asymptotic of discounted aggregate claims but the upper tail asymptotic independence is negligible.     

A class of partially linear transformation models for recurrent gap times

In this article, we propose a general class of partially linear transformation models for recurrent gap time data, which extends the linear transformation models by incorporating non linear covariate effects and includes the partially linear proportional hazards and the partially linear proportional odds models as special cases. Both global and local estimating equations are developed to estimate the parametric and non parametric covariate effects, and the asymptotic properties of the resulting estimators are established. The finite-sample behavior of the proposed estimators is evaluated through simulation studies, and an application to a clinic study on chronic granulomatous disease is provided.     

The beta generalized linear exponential distribution

In this paper, a new five-parameter lifetime distribution called beta generalized linear exponential distribution (BGLED) is introduced. It includes at least 17 popular sub-models as special cases such as the beta linear exponential, the beta generalized exponential, and the exponentiated generalized linear distributions. Mathematical and statistical properties of the proposed distribution are discussed in details. In particular, explicit expression for the density function, moments, asymptotics distributions for sample extreme statistics, and other statistical measures are obtained. The estimation of the parameters by the method of maximum-likelihood is discussed and the finite sample properties of the maximum-likelihood estimators (MLEs) are investigated numerically. A real data set is used to demonstrate its superior performance fit over several existing popular lifetime models.     

Quasi-minimax estimation in the partial linear model

This article discusses the minimax estimator in partial linear model y = Zβ + f + ε under ellipsoidal restrictions on the parameter space and quadratic loss function. The superiority of the minimax estimator over the two-step estimator is studied in the mean squared error matrix criterion.     

Why do we observe misclassification errors smaller than the Bayes error?

In simulation studies for discriminant analysis, misclassification errors are often computed using the Monte Carlo method, by testing a classifier on large samples generated from known populations. Although large samples are expected to behave closely to the underlying distributions, they may not do so in a small interval or region, and thus may lead to unexpected results. We demonstrate with an example that the LDA misclassification error computed via the Monte Carlo method may often be smaller than the Bayes error. We give a rigorous explanation and recommend a method to properly compute misclassification errors.     

Strategies for fostering basic psychological needs support in high quality youth leadership programs

Youth leadership programming has become an increasingly common context to foster basic psychological needs and promote youth development. The purpose of this qualitative study was to explore strategies involved in fostering youth needs support within six leadership programs. Two leaders and 30 youth participated in semi-structured interviews to better understand the strategies used to foster needs support. Findings revealed that leaders were able to foster a sense of relatedness among youth through building trusting adult-youth relationships and nurturing an inclusive environment. Maximizing choice and negotiating youth voice helped to foster youth’s autonomy. Finally, creating a task-oriented climate and providing intentional opportunities for skill-building helped to foster youth’s competence. Findings suggest that training for leaders is critical in understanding what, and how strategies should be employed to help foster youth needs support in leadership programming. Limitations and future directions are outlined.     

Optimizing global thermal coal shipments

Thermal coal is used to produce energy; with changing emissions standards and advents in renewable technology, the thermal coal market has seen significant transformation over the past decade. We develop a mixed-integer optimization problem that seeks to minimize shipment costs while meeting demand for thermal coal, and which respects quality constraints, supply limits, and port capacity; we use this model to analyze the following scenarios: (i) a counterfactual setting in which we compare historical shipping patterns to model results using a 2012 base year; (ii) the explicit effect of Chinese mandates on coal shipments; (iii) the impact on our shipping patterns of reduced Chinese and Indian demand; (iv) the effects of the Baltic Dry Index and oil prices; and (v) a comparison of shipments prior and subsequent to Panama Canal expansion. Our work can be used to inform policy, study responses to variable price and demand scenarios, and provide insight to both coal producers and consumers about the international coal market. For example, removal of mandates set by the Chinese government to fill its own demand decreases coal flows from Northern to Southern China by 56%, which has a spill-over effect on European and American markets; and, expansion of the Panama Canal leads to only modest shipping increases through the canal (6.7%), with more coal originating from Colombia serving Asian demand.     

La répartition régionale de la prime aux qualifications en Chine urbaine: implications sur la croissance et les inégalités

Les auteurs observent que la prime aux qualifications a augmenté dans toutes les régions de Chine entre 1995 et 2002, mais seulement dans les régions côtières entre 2002 et 2007. Ainsi, ces régions présentent de plus fortes inégalités salariales, notamment en zones urbaines. Si les privatisations ont été la première raison de la montée de la prime aux qualifications entre 1995 et 2002, entre 2002 et 2007 cela a été l'intégration, inégale selon les régions, à l'économie mondiale. Réduire cette prime et les inégalités suppose de réformer le système du hukou qui fait obstacle à la mobilité de la main‐d'?uvre, donc à la croissance.