Bartlett corrections in beta regression models

We consider the issue of performing accurate small-sample testing inference in beta regression models, which are useful for modeling continuous variates that assume values in (0,1), such as rates and proportions. We derive the Bartlett correction to the likelihood ratio test statistic and also consider a bootstrap Bartlett correction. Using Monte Carlo simulations we compare the finite sample performances of the two corrected tests to that of the standard likelihood ratio test and also to its variant that employs Skovgaard's adjustment; the latter is already available in the literature. The numerical evidence favors the corrected tests we propose. We also present an empirical application.     

Heteroskedasticity-Robust Inference in Linear Regressions

The assumption that all errors share the same variance (homoskedasticity) is commonly violated in empirical analyses carried out using the linear regression model. A widely adopted modeling strategy is to perform point estimation by ordinary least squares and then perform testing inference based on these point estimators and heteroskedasticity-consistent standard errors. These tests, however, tend to be size-distorted when the sample size is small and the data contain atypical observations. Furno (1996 Furno , M. ( 1996 ). Small sample behavior of a robust heteroskedasticity consistent covariance matrix estimator . Journal of Statistical Computation and Simulation 54 : 115 – 128 .[Taylor & Francis Online], [Web of Science ®] , [Google Scholar]) suggested performing point estimation using a weighted least squares mechanism in order to attenuate the effect of leverage points on the associated inference. In this article, we follow up on her proposal and define heteroskedasticity-consistent covariance matrix estimators based on residuals obtained using robust estimation methods. We report Monte Carlo simulation results (size and power) on the finite sample performance of different heteroskedasticity-robust tests. Overall, the results favor inference based on HC0 tests constructed using robust residuals.     

An Experiment on Rational Insurance Decisions

We describe the results of an experiment on decision making in an insurance context. The experiment was designed to test for the underlying rationality of insurance consumers, where rationality is understood in usual economic terms. In particular, using expected utility as the preference function, we test for positive marginal utility, risk aversion, and decreasing absolute risk aversion, all of which are normal postulates for any microeconomic decision context under uncertainty or risk. We find that there the discrepancy from rational decision making increases with the sophistication of the rationality criteria, that irrationality concerning fair premium contracts is uncharacteristically high, and that the slope of absolute risk aversion seems to depend on the format of the insurance contract. This revised version was published online in June 2006 with corrections to the Cover Date.     

The effect of reparametrization on accelerated lifetime tests

Efficient reliability industrial experiments consist of submitting items to accelerated life tests. It is of interest to obtain measures of the realiability of the devices under the usual working conditions, represented here by the mean lifetime. A practical problem refers to the accuracy of interval estimation of the parameter of interest when the sample size is small or moderate. In this paper, we describe the effect of several reparametrizations on the accuracy of the interval estimation. We propose a reparametrization that leads to accuracy while allowing orthogonality between the parameters. The idea is to consider a logarithmic reparametrization on orthogonal parameters in order to have independent maximum likelihood estimates with good asymptotic normal approximation. The study is illustrated by a data set on an accelerated life test at pressurized containers of Kevlan/Epoxy 49.     

Rescuing gender: An exploration of embodied resilience-making in the Chilean miners' catastrophe

In this paper, we explore the theoretical and empirical links between gender and resilience within the context of crisis management by developing a theoretically oriented analysis of the iconic case of the 33 Chilean miners rescue. Analytic focus is placed on the capacity of resilience-promoting actors to embody and transform the discursive and practical dimensions of gendered regimes that predominate at an organizational level. As a result, we can distinguish four distinct modalities—normative redefinition, emotional reaccommodation, resistance, and affective affirmation—through which the gendered dimension of resilience becomes undone and redone, based on the action of alternative rationalities and affects. Overall, our findings indicate that resilience cannot be made possible without effectively disputing the norms that dispose how bodies, in all their diversity, are able to be recognized and socially experienced to face and manage crises effectively.     

Macroeconomic Determinants of Equilibrium Unemployment Insurance

This paper proposes a new modeling strategy as regards the definition of an optimal level of unemployment benefits. While the traditional methodology privileges labor market equilibrium to derive optimal employment, wage and unemployment benefit levels, we present a model in which the optimal level of unemployment benefits is a function of the government’s macroeconomic objectives in terms of inflation and output fluctuations. In a second stage, the model allows for the investigation of unemployment insurance effects on labor market equilibrium.     

A new heteroskedasticity-consistent covariance matrix estimator for the linear regression model

The assumption that all random errors in the linear regression model share the same variance (homoskedasticity) is often violated in practice. The ordinary least squares estimator of the vector of regression parameters remains unbiased, consistent and asymptotically normal under unequal error variances. Many practitioners then choose to base their inferences on such an estimator. The usual practice is to couple it with an asymptotically valid estimation of its covariance matrix, and then carry out hypothesis tests that are valid under heteroskedasticity of unknown form. We use numerical integration methods to compute the exact null distributions of some quasi-t test statistics, and propose a new covariance matrix estimator. The numerical results favor testing inference based on the estimator we propose.     

Risk ranking of food categories associated with Salmonella enterica contamination in the central region of Mexico

To prevent and control foodborne diseases, there is a fundamental need to identify the foods that are most likely to cause illness. The goal of this study was to rank 25 commonly consumed food products associated with Salmonella enterica contamination in the Central Region of Mexico. A multicriteria decision analysis (MCDA) framework was developed to obtain an S. enterica risk score for each food product based on four criteria: probability of exposure to S. enterica through domestic food consumption (Se); S. enterica growth potential during home storage (Sg); per capita consumption (Pcc); and food attribution of S. enterica outbreak (So). Risk scores were calculated by the equation Se*W₁+Sg*W₂+Pcc*W₃+So*W₄, where each criterion was assigned a normalized value (1–5) and the relative weights (W) were defined by 22 experts’ opinion. Se had the largest effect on the risk score being the criterion with the highest weight (35%; IC95% 20%–60%), followed by So (24%; 5%–50%), Sg (23%; 10%–40%), and Pcc (18%; 10%–35%). The results identified chicken (4.4 ± 0.6), pork (4.2 ± 0.6), and beef (4.2 ± 0.5) as the highest risk foods, followed by seed fruits (3.6 ± 0.5), tropical fruits (3.4 ± 0.4), and dried fruits and nuts (3.4 ± 0.5), while the food products with the lowest risk were yogurt (2.1 ± 0.3), chorizo (2.1 ± 0.4), and cream (2.0 ± 0.3). Approaches with expert-based weighting and equal weighting showed good correlation (R² = 0.96) and did not show significant differences among the ranking order in the top 20 tier. This study can help risk managers select interventions and develop targeted surveillance programs against S. enterica in high-risk food products.     

How to Code a Million Missions: Developing Bespoke Nonprofit Activity Codes Using Machine Learning Algorithms

VOLUNTAS: International Journal of Voluntary and Nonprofit Organizations - National Taxonomy of Exempt Entities (NTEE) codes have become the primary classifier of nonprofit missions since they were...