A class of bivariate regression models for discrete and/or continuous responses

ABSTRACT

A general class of models for discrete and/or continuous responses is proposed in which joint distributions are constructed via the conditional approach. It is assumed that the distributions of one response and of the other response given the first one belong to exponential family of distributions. Furthermore, the marginal means are related to the covariates by link functions and a dependency structure between the responses is inserted into the model. Estimation methods, diagnostic analysis and a simulation study considering a Bernoulli-exponential model, a particular case of the class, are presented. Finally, this model is used in a real data set.     

Geometry in Canonical Correlations

In this article, a useful proposition relating the canonical correlations in multi-way layout to the singular values of a specific matrix is proved and also a geometrical explanation of canonical correlations as an angle between subspaces is given.     

Measuring Kurtosis by Right and Left Inequality Orders

The question of how to measure kurtosis in both symmetric and asymmetric distributions is addressed using the kurtosis diagram of Zenga (2006 Zenga , M. ( 2006 ). Kurtosis . In : Kotz , S. , Read , C. B. , Balakrishnan , N. , Vidakovic , B. , eds. Encyclopedia of Statistical Sciences. 2nd ed . New York : John Wiley and Sons . [Google Scholar]). Kurtosis is related to inequality at either side of the median, and we establish a hierarchy of kurtosis orderings in which the kurtosis diagram stands at the weakest level. A sufficient condition for constructing kurtosis measures compatible with such ordering is provided. The merits of the proposed approach in both clarifying and formalizing the idea of kurtosis are evaluated and examples are discussed throughout.     

Testing citizen identities: Australian migrants and the Australian values debate

In this paper I explore how recent public debates about Australian values impact on the everyday lives of migrants preparing to become naturalised Australian citizens. Framed through the perspective of governmentality, I combine textual analyses of government documents relating to the Australian citizenship test with ethnographies of recently-arrived migrants in order to demonstrate how political practices related to the conferral of Australian citizenship link and interconnect with the production of migrant subjectivities. In this way, I explore what aspiring citizens think about Australian values and how they contest, accept and negotiate these demands on them to adopt the Australian way of life. The analysis suggests that migrants in this study presented multiple and complex ways of being model Australian citizens. They drew on popular national myths about Australianness promoted in citizenship test resources but they also produced alternative subjectivities that promoted identification with everyday multiculturalism as central to living a happy and prosperous life in Australia.     

20 years of Bologna - a story of success,a story of failure

The Bologna Process is a unique harmonisation process taking place outside the policy-making framework of the European Union. It aims at enhancing the comparability and compatibility of higher education structures and degrees across Europe, as well as to institutionalise quality assurance mechanisms. The aim of this article is to provide a condensed, up-to-date overview of the Bologna Process with regard to structural characteristics, before embedding it into a discussion on processes on voluntary policy convergence and to which extent we should be able to find this kind of policy harmonisation in the realm of the Bologna Process. Related to this are questions why this, in principle, completely voluntary process of policy harmonisation, has appealed to so many countries and why they might or might not feel committed to the implementation of its policies and tools.     

Testing Overdispersion in CUBE Models

A practical problem with large-scale survey data is the possible presence of overdispersion. It occurs when the data display more variability than is predicted by the variance–mean relationship. This article describes a probability distribution generated by a mixture of discrete random variables to capture uncertainty, feeling, and overdispersion. Specifically, several tests for detecting overdispersion will be implemented on the basis of the asymptotic theory for maximum likelihood estimators. We discuss the results of a simulation experiment concerning log-likelihood ratio, Wald, Score, and Profile tests. Finally, some real datasets are analyzed to illustrate the previous results.     

Influence diagnostics for the Weibull-Negative-Binomial regression model with cure rate under latent failure causes

In this paper, we propose a flexible cure rate survival model by assuming that the number of competing causes of the event of interest follows the Negative Binomial distribution and the time to event follows a Weibull distribution. Indeed, we introduce the Weibull-Negative-Binomial (WNB) distribution, which can be used in order to model survival data when the hazard rate function is increasing, decreasing and some non-monotonous shaped. Another advantage of the proposed model is that it has some distributions commonly used in lifetime analysis as particular cases. Moreover, the proposed model includes as special cases some of the well-know cure rate models discussed in the literature. We consider a frequentist analysis for parameter estimation of a WNB model with cure rate. Then, we derive the appropriate matrices for assessing local influence on the parameter estimates under different perturbation schemes and present some ways to perform global influence analysis. Finally, the methodology is illustrated on a medical data.     

Investigating impacts of complex sampling on latent growth curve modelling

We investigate the impacts of complex sampling on point and standard error estimates in latent growth curve modelling of survey data. Methodological issues are illustrated with empirical evidence from the analysis of longitudinal data on life satisfaction trajectories using data from the British Household Panel Survey, a national representative survey in Great Britain. A multi-process second-order latent growth curve model with conditional linear growth is used to study variation in the two perceived life satisfaction latent factors considered. The benefits of accounting for the complex survey design are considered, including obtaining unbiased both point and standard error estimates, and therefore correctly specified confidence intervals and statistical tests. We conclude that, even for the rather elaborated longitudinal data models that were considered, estimation procedures are affected by variance-inflating impacts of complex sampling.