On the mathematics of the Jeffreys–Lindley paradox

This paper is concerned with the well known Jeffreys–Lindley paradox. In a Bayesian set up, the so-called paradox arises when a point null hypothesis is tested and an objective prior is sought for the alternative hypothesis. In particular, the posterior for the null hypothesis tends to one when the uncertainty, i.e., the variance, for the parameter value goes to infinity. We argue that the appropriate way to deal with the paradox is to use simple mathematics, and that any philosophical argument is to be regarded as irrelevant.     

A note on the posterior inference for the Yule–Simon distribution

The Yule–Simon distribution has been out of the radar of the Bayesian community, so far. In this note, we propose an explicit Gibbs sampling scheme when a Gamma prior is chosen for the shape parameter. The performance of the algorithm is illustrated with simulation studies, including count data regression, and a real data application to text analysis. We compare our proposal to the frequentist counterparts showing better performance of our algorithm when a small sample size is considered.     

LEANING AGAINST WINDY BANK LENDING

Using an estimated dynamic stochastic general equilibrium model with banking, this paper first provides evidence that monetary policy reacted to bank loan growth in the United States during the Great Moderation. It then shows that the optimized simple interest‐rate rule features no response to the growth of bank credit. However, the welfare loss associated to the empirical responsiveness is small. The sources of business cycle fluctuations are crucial in determining whether a “leaning‐against‐the‐wind” policy is optimal or not. In fact, the predominant role of supply shocks in the model gives rise to a trade‐off between inflation and financial stabilization. (JEL E32, E44, E52)     

A non-parametric regression approach to repeated measures analysis in cancer experiments

The validity conditions for univariate or multivariate analyses of repeated measures are highly sensitive to the usual assumptions. In cancer experiments, the data are frequently heteroscedastic and strongly correlated with time, and standard analyses do not perform well. Alternative non-parametric approaches can contribute to an analysis of these longitudinal data. This paper describes a method for such situations, using the results from a comparative experiment in which tumour volume is evaluated over time. First, we apply the non-parametric approach proposed by Raz in constructing a randomization Ftest for comparing treatments. A local polynomial fit is conducted to estimate the growth curves and confidence intervals for each treatment. Finally, this technique is used to estimate the velocity of tumour growth.     

Symbolische Gewalt und ihr potenzielles Scheitern. Eine Ann?herung zwischen Butler und Bourdieu

The article discusses the similarities and differences between the contributions of Pierre Bourdieu and Judith Butler regarding symbolic power. While both share core arguments such as the discursive constitution of symbolic power, its disguised effectiveness due to naturalization processes, and the notion that symbolic power is part of domination (as it is interwoven with the social logic of personal recognition), there are also important differences between Butler and Bourdieu. Latter insists on discerning the social dimension of discourse??asking who speaks from which position???from the immanent linguistic realm of discourse. Butler instead highlights the impossibility of separating the social from the discursive and stresses the imponderability of discursive acts due to their performativity and iterative logic. Thus, Butler argues, ??social imperatives?? (Bourdieu) can and actually do ??fail??. This becomes especially evident within processes of subjectivation (Butler). Considering the systematic possibility of ??failure?? in Butlers work, I argue that Butler and Bourdieu offer distinct notions of critique: While Butler??s notion is framed as heterotopy (Foucault), Bourdieu??s is utopic.     

Loss-based approach to two-piece location-scale distributions with applications to dependent data

Two-piece location-scale models are used for modeling data presenting departures from symmetry. In this paper, we propose an objective Bayesian methodology for the tail parameter of two particular distributions of the above family: the skewed exponential power distribution and the skewed generalised logistic distribution. We apply the proposed objective approach to time series models and linear regression models where the error terms follow the distributions object of study. The performance of the proposed approach is illustrated through simulation experiments and real data analysis. The methodology yields improvements in density forecasts, as shown by the analysis we carry out on the electricity prices in Nordpool markets.

     

An Objective Bayesian Criterion to Determine Model Prior Probabilities

We discuss the problem of selecting among alternative parametric models within the Bayesian framework. For model selection problems, which involve non‐nested models, the common objective choice of a prior on the model space is the uniform distribution. The same applies to situations where the models are nested. It is our contention that assigning equal prior probability to each model is over simplistic. Consequently, we introduce a novel approach to objectively determine model prior probabilities, conditionally, on the choice of priors for the parameters of the models. The idea is based on the notion of the worth of having each model within the selection process. At the heart of the procedure is the measure of this worth using the Kullback–Leibler divergence between densities from different models.     

Kidnap risks and migration: evidence from Colombia

Using a unique data set from the major Colombian cities collected between 2000–2003 and with information on more than 12,000 households, this paper studies the relationship between the kidnap risk a household faces with its migration decisions. We find evidence that exposure to such risk induces households to react sending some of their members to an international destination but not necessarily to a domestic one. Estimates are robust to the inclusion of several household characteristics usual in the migration literature, other crime risks, reported feelings of insecurity of the household, and an alternative measure of kidnap risk.