Bayesian Longitudinal Data Analysis with Mixed Models and Thick-tailed Distributions using MCMC

Linear mixed effects models are frequently used to analyse longitudinal data, due to their flexibility in modelling the covariance structure between and within observations. Further, it is easy to deal with unbalanced data, either with respect to the number of observations per subject or per time period, and with varying time intervals between observations. In most applications of mixed models to biological sciences, a normal distribution is assumed both for the random effects and for the residuals. This, however, makes inferences vulnerable to the presence of outliers. Here, linear mixed models employing thick-tailed distributions for robust inferences in longitudinal data analysis are described. Specific distributions discussed include the Student-t, the slash and the contaminated normal. A Bayesian framework is adopted, and the Gibbs sampler and the Metropolis-Hastings algorithms are used to carry out the posterior analyses. An example with data on orthodontic distance growth in children is discussed to illustrate the methodology. Analyses based on either the Student-t distribution or on the usual Gaussian assumption are contrasted. The thick-tailed distributions provide an appealing robust alternative to the Gaussian process for modelling distributions of the random effects and of residuals in linear mixed models, and the MCMC implementation allows the computations to be performed in a flexible manner.     

Etherapy: A Training Program for Development of Clinical Skills in Distance Psychotherapy

ABSTRACT

This article describes a project by the Virtual Teaching and Cyberpsychology Laboratory of the National Autonomous University of Mexico developed to (1) implement a teaching program for clinical psychology students to learn therapeutic skills in providing online psychological services; (2) evaluate the students' acquisition of knowledge and skills in diagnosis via the Internet, case formulation of treatment, and adherence to treatment protocols; and (3) incorporate technologies for creating training settings and online professional services in order to develop the students' clinical competency in this treatment modality. A pilot study was conducted with 17 clinical psychology students in the first phase of the teaching program and 6 students during the second phase. Preliminary results indicate that students increased knowledge of Internet-based clinical interventions and improved in clinical skills. It is necessary to continue the evaluation of the teaching program with larger samples, and further research is suggested.     

Mixed integer programming formulations for clustering problems related to structural balance

In this work, we study graph clustering problems associated with structural balance. One of these problems is known in computer science literature as the correlation-clustering (CC) problem and another (RCC) can be viewed as its relaxed version. The solution of CC and RCC problems has been previously used in the literature as tools for the evaluation of structural balance in a social network. Our aim is to solve these problems to optimality. We describe integer linear programming formulations for these problems which includes the first mathematical formulation for the RCC problem. We also discuss alternative models for the relaxed structural balance and the solution of clustering problems associated with these new models. Numerical experiments are carried out with each formulation on a set of benchmark instances available in the literature.     

The decreasing percentile residual life ageing notion

Earlier researchers have studied some aspects of the classes of distribution functions with decreasing α-percentile residual life (DPRL(α)), 0<α<1. The purpose of this paper is to note some further properties of these classes, and to initiate a theory of non-parametric statistical estimation of DPRL(α) functions. Specifically, the close relationship between the DPRL(α) and the increasing failure rate ageing notions is studied. Other close relationships, between the DPRL(α) ageing notions and the percentile residual life stochastic orders, are described, and further properties of the above classes of distributions are derived. Finally, we introduce an estimator of the percentile residual life function, under the condition that it decreases, and we prove its strongly uniform consistency.     

Simulation analysis of cold chain performance based on time–temperature data

Perishable goods are a fundamental source of revenue for the retail sector; their management, however, constitutes a severe challenge for retailers and supply chain partners. A significant cost in particular is the fraction of products perished through the supply chain, which also constitutes an ethical and environmental concern. Supply chain organisation and operative characteristics have a significant influence on this matter, as in fact ensuring suitable temperature conditions for the stock-keeping units throughout the supply chain is mandatory for perishable products. Recent developments in sensing and communication technologies allow detailed monitoring and control of cold chain; however, depending on the characteristics of the supply chain, an inherent risk of perished products is often inevitable, even in the hypothesis of perfect control. This article proposes a methodology to evaluate the performance of a cold chain in terms of expected product quality at the retail store, and to estimate the expected fraction of perished products, according to the supply chain configuration. The approach is based on Monte Carlo simulation, and implements referenced shelf-life models. A real application is also presented, involving a preliminary analysis and mapping of the supply chain activities based on time–temperature data, in order to demonstrate the practicability of the approach proposed.     

PROMETHEE IV through kernel density estimation

ABSTRACT

This paper presents a new version of PROMETHEE IV, which considers the empirical distribution of the criteria through kernel density estimation to evaluate alternatives. The developed method has the ability to treat criteria according to their distribution. The classic PROMETHEE IV can produce divergent integrals, and this could be the cause for its insufficient exploration in literature. The proposed method overcomes this situation since large values have little weight compared to values near the mean.     

ATTRIBUTION ERROR IN ECONOMIC VOTING: EVIDENCE FROM TRADE SHOCKS

This article exploits the international transmission of business cycles to examine the prevalence of attribution error in economic voting in a large panel of countries from 1990 to 2009. We find that voters, on average, exhibit a strong tendency to oust the incumbent governments during an economic downturn, regardless of whether the recession is home‐grown or merely imported from trading partners. However, we find important heterogeneity in the extent of attribution error. A split sample analysis shows that countries with more experienced voters, more educated voters, and possibly more informed voters—all conditions that have been shown to mitigate other voter agency problems—do better in distinguishing imported from domestic growth. (JEL E3, E6)