Bayesian robustness of the compound Poisson distribution under bidimensional prior: an application to the collective risk model

The distribution of the aggregate claims in one year plays an important role in Actuarial Statistics for computing, for example, insurance premiums when both the number and size of the claims must be implemented into the model. When the number of claims follows a Poisson distribution the aggregated distribution is called the compound Poisson distribution. In this article we assume that the claim size follows an exponential distribution and later we make an extensive study of this model by assuming a bidimensional prior distribution for the parameters of the Poisson and exponential distribution with marginal gamma. This study carries us to obtain expressions for net premiums, marginal and posterior distributions in terms of some well-known special functions used in statistics. Later, a Bayesian robustness study of this model is made. Bayesian robustness on bidimensional models was deeply treated in the 1990s, producing numerous results, but few applications dealing with this problem can be found in the literature.     

Concentration function and sensitivity to the prior

Summary In robust bayesian analysis, ranges of quantities of interest (e. g. posterior means) are usually considered when the prior probability measure varies in a class Γ. Such quantities describe the variation of just one aspect of the posterior measure. The concentration function describes changes in the posterior probability measure more globally, detecting differences in probability concentration and providing, simultaneously, bounds on the posterior probability of all measurable subsets. In this paper, we present a novel use of the concentration function, and two concentration indices, to study such posterior changes for a general class Γ, restricting then our attention to some ∈-contamination classes of priors.     

Block unimodality for multivariate Bayesian robustness

Summary The development of Bayesian robustness has been growing in the last decade. The theory has extensively dealt with the univariate parameter case. Among the vast amount of proposals in the literature, only a few of them have a straightforward extension to the multivariate case. In this paper we consider the multidimensional version of the class of ε-contaminated prior distributions, with unimodal contaminations. In the multivariate case there is not a unique definition of unimodality and one's choice must be based on statistical ground. Here we propose the use of the block unimodal distributions, which proved to be very suitable for modelling situations where the coordinates of the parameter ϑ are deemed, a priori, weakly correlated.     

Model influence functions based on mixtures

Influence functions are considered as diagnostics for model departures in parametric Bayesian inference. A baseline model density is expressed as a mixture; then the mixing distribution is perturbed. This is designed to engender perturbations which are plausible a priori. The influence of perturbations is measured for both Bayes estimates and their associated posterior expected losses. To assess the plausibility of perturbations a posteriori, an additional influence function is constructed for the Bayes factor comparing the perturbed and baseline models. The effect of perturbation on various estimands is incorporated in the analysis.     

An approach to evaluating sensitivity in Bayesian regression analyses

This paper presents a method for assessing the sensitivity of predictions in Bayesian regression analyses. In parametric Bayesian analyses there is a family s₀ of regression functions, parametrized by a finite-dimensional vector B. The family s₀ is a subset of R, the set of all possible regression functions. A prior π₀ on B induces a prior on R. This paper assesses sensitivity by computing bounds on the predictive probability of a fixed set K over a class of priors, Γ, induced by a class of families of regression functions, Γ_s, and a class of priors, Γ_π. This paper is divided into three parts which (1) define Γ, (2) describe an algorithm for finding accurate bounds on predictive probabilities over Γ and (3) illustrate the method with two examples. It is found that sensitivity to the family of regression functions can be much more important than sensitivity to π₀.     

Examples of doubly stochastic measures supported on the graphs of two functions

A doubly stochastic measure (DSM) is a measure μ on the unit square so that μ([0, 1] × A) = μ(A × [0, 1]) = m(A) where m is Lebesgue measure. The set of DSMs forms a convex set in the space of measures. It is known that DSMs supported on the union of two graphs of invertible functions are extreme points of that convex set (Seethoff and Shiflett, 1977/78 Seethoff, T.L., Shiflett. (1977/78). Doubly stochastic measures with prescribed support. Z. Wahrscheinlichkeitstheorie und Verw. Gebiete 41(4):283–288.[Crossref], [Web of Science ®] , [Google Scholar]). In general, there are few examples of extreme points in the literature. There are examples of so-called hairpins where the functions involved are inverses of each other, but there are also examples of the union of the graphs of a function and its inverse does not support a DSM (Sherwood and Taylor, 1988 Sherwood, H., Taylor, M.D. (1988). Doubly stochastic measures with hairpin support. Probab. Theory Related Fields 78(4):617–626.[Crossref], [Web of Science ®] , [Google Scholar]). In this paper, for a function f in a certain class, we find companion functions g so that the union of the graphs of f and g support a DSM even though the union of the graphs of f and f-inverse do not.     

Bayesian likelihood robustness in linear models

This paper deals with the problem of robustness of Bayesian regression with respect to the data. We first give a formal definition of Bayesian robustness to data contamination, prove that robustness according to the definition cannot be obtained by using heavy-tailed error distributions in linear regression models and propose a heteroscedastic approach to achieve the desired Bayesian robustness.     

Exact Bayesian modeling for bivariate Poisson data and extensions

Bivariate count data arise in several different disciplines (epidemiology, marketing, sports statistics just to name a few) and the bivariate Poisson distribution being a generalization of the Poisson distribution plays an important role in modelling such data. In the present paper we present a Bayesian estimation approach for the parameters of the bivariate Poisson model and provide the posterior distributions in closed forms. It is shown that the joint posterior distributions are finite mixtures of conditionally independent gamma distributions for which their full form can be easily deduced by a recursively updating scheme. Thus, the need of applying computationally demanding MCMC schemes for Bayesian inference in such models will be removed, since direct sampling from the posterior will become available, even in cases where the posterior distribution of functions of the parameters is not available in closed form. In addition, we define a class of prior distributions that possess an interesting conjugacy property which extends the typical notion of conjugacy, in the sense that both prior and posteriors belong to the same family of finite mixture models but with different number of components. Extension to certain other models including multivariate models or models with other marginal distributions are discussed.     

A Bayesian method for classification and discrimination

We discuss Bayesian analyses of traditional normal-mixture models for classification and discrimination. The development involves application of an iterative resampling approach to Monte Carlo inference, commonly called Gibbs sampling, and demonstrates routine application. We stress the benefits of exact analyses over traditional classification and discrimination techniques, including the ease with which such analyses may be performed in a quite general setting, with possibly several normal-mixture components having different covariance matrices, the computation of exact posterior classification probabilities for observed data and for future cases to be classified, and posterior distributions for these probabilities that allow for assessment of second-level uncertainties in classification.     

Bayesian robustness in bidimensional models: Prior independence

When θ is a multidimensional parameter, the issue of prior dependence or independence of coordinates is a serious concern. This is especially true in robust Bayesian analysis; Lavine et al. (J. Amer. Statist. Assoc.86, 964–971 (1991)) show that allowing a wide range of prior dependencies among coordinates can result in near vacuous conclusions. It is sometimes possible, however, to make confidently the judgement that the coordinates of θ are independent a priori and, when this can be done, robust Bayesian conclusions improve dramatically. In this paper, it is shown how to incorporate the independence assumption into robust Bayesian analysis involving -contamination and density band classes of priors. Attention is restricted to the case θ = (θ₁, θ₂) for clarity, although the ideas generalize.     

On the robustness to outliers of the Student-t process

The theory of Bayesian robustness modeling uses heavy-tailed distributions to resolve conflicts of information by rejecting automatically the outlying information in favor of the other sources of information. In particular, the Student's-t process is a natural alternative to the Gaussian process when the data might carry atypical information. Several works attest to the robustness of the Student $t$ process, however, the studies are mostly guided by intuition and focused mostly on the computational aspects rather than the mathematical properties of the involved distributions. This work uses the theory of regular variation to address the robustness of the Student $t$ process in the context of nonlinear regression, that is, the behavior of the posterior distribution in the presence of outliers in the inputs, in the outputs, or in both sources of information. In all these cases, under certain conditions, it is shown that the posterior distribution tends to a quantity that does not depend on the atypical information, then, for every case, the limiting posterior distribution as the outliers tend to infinity is provided. The impact of outliers on the predictive posterior distribution is also addressed. The theory is illustrated with a few simulated examples.     

On Bayesian learning via loss functions

We provide a decision theoretic approach to the construction of a learning process in the presence of independent and identically distributed observations. Starting with a probability measure representing beliefs about a key parameter, the approach allows the measure to be updated via the solution to a well defined decision problem. While the learning process encompasses the Bayesian approach, a necessary asymptotic consideration then actually implies the Bayesian learning process is best. This conclusion is due to the requirement of posterior consistency for all models and of having standardized losses between probability distributions. This is shown considering a specific continuous model and a very general class of discrete models.     

A study of Bayesian local robustness with applications in actuarial statistics

Local or infinitesimal Bayesian robustness is a powerful tool to study the sensitivity of posterior magnitudes, which cannot be expressed in a simple manner. For these expressions, the global Bayesian robustness methodology does not seem adequate since the practitioner cannot avoid using inappropriate classes of prior distributions in order to make the model mathematically tractable. This situation occurs, for example, when we compute some types of premiums in actuarial statistics in order to fix the premium to be charged to an insurance policy. In this paper, analytical and simple expressions that allow us to study the sensitivity of premiums, which are usually used in automobile insurance are provided by using the local Bayesian robustness methodology. Some examples are examined by using real automobile claim insurance data.     

Measuring robustness for weighted distributions: Bayesian perspective

There are many situations where the usual random sample from a population of interest is not available, due to the data having unequal probabilities of entering the sample. The method of weighted distributions models this ascertainment bias by adjusting the probabilities of actual occurrence of events to arrive at a specification of the probabilities of the events as observed and recorded. We consider two different classes of contaminated or mixture of weight functions, Γ _a ={w(x):w(x)=(1−ε)w ₀(x)+εq(x),q∈Q} and Γ _g ={w(x):w(x)=w ₀ ^1−ε (x)q ^ε(x),q∈Q} wherew ₀(x) is the elicited weighted function,Q is a class of positive functions and 0≤ε≤1 is a small number. Also, we study the local variation of ϕ-divergence over classes Γ _a and Γ _g . We devote on measuring robustness using divergence measures which is based on the Bayesian approach. Two examples will be studied.     

The Non Dominated Set in Bayesian Decision Problems with Convex Loss Functions

Bayesian decision problems require subjective elicitation of the inputs: beliefs and preferences. Sometimes, elicitation methods may not represent perfectly the judgements of the decision maker. Several foundations propose to overlay this problem using robust approaches. In these models, beliefs are modelled by a class of probability distributions and preferences by a class of loss functions. Then, we are in the conditions of a Pareto order. Hence the solution concept is the set of non dominated alternatives. In this article we focus on the computation of the efficient set when the preferences are modeled by a class of convex loss functions.     

Robust Bayesian analysis of loss reserving data using scale mixtures distributions

It is vital for insurance companies to have appropriate levels of loss reserving to pay outstanding claims and related settlement costs. With many uncertainties and time lags inherently involved in the claims settlement process, loss reserving therefore must be based on estimates. Existing models and methods cannot cope with irregular and extreme claims and hence do not offer an accurate prediction of loss reserving. This paper extends the conventional normal error distribution in loss reserving modeling to a range of heavy-tailed distributions which are expressed by certain scale mixtures forms. This extension enables robust analysis and, in addition, allows an efficient implementation of Bayesian analysis via Markov chain Monte Carlo simulations. Various models for the mean of the sampling distributions, including the log-Analysis of Variance (ANOVA), log-Analysis of Covariance (ANCOVA) and state space models, are considered and the straightforward implementation of scale mixtures distributions is demonstrated using OpenBUGS.     

Bayesian Analysis of Least Absolute Relative Error Regression

In an important article by Chen et al. (2010) introduced a new distribution compatible with maximum likelihood estimation in a Least Absolute Relative Error (LARE) setting. In this article, we show first that the posterior of the model is log – concave and thus specialized and highly efficient techniques can be used to perform Bayesian inference without the use of MCMC since they provide independent draws from the posterior. Second, we approximate the distribution using a finite mixture of normals. Surprisingly, the log-LARE distribution can be approximated using a finite scale mixture of normals with few components.     

Bayesian quadrature with non-normal approximating functions

We consider an efficient Bayesian approach to estimating integration-based posterior summaries from a separate Bayesian application. In Bayesian quadrature we model an intractable posterior density function f(·) as a Gaussian process, using an approximating function g(·), and find a posterior distribution for the integral of f(·), conditional on a few evaluations of f (·) at selected design points. Bayesian quadrature using normal g (·) is called Bayes-Hermite quadrature. We extend this theory by allowing g(·) to be chosen from two wider classes of functions. One is a family of skew densities and the other is the family of finite mixtures of normal densities. For the family of skew densities we describe an iterative updating procedure to select the most suitable approximation and apply the method to two simulated posterior density functions.     

Bayesian Mixture Labeling and Clustering

Label switching is one of the fundamental issues for Bayesian mixture modeling. It occurs due to the nonidentifiability of the components under symmetric priors. Without solving the label switching, the ergodic averages of component specific quantities will be identical and thus useless for inference relating to individual components, such as the posterior means, predictive component densities, and marginal classification probabilities. The author establishes the equivalence between the labeling and clustering and proposes two simple clustering criteria to solve the label switching. The first method can be considered as an extension of K-means clustering. The second method is to find the labels by minimizing the volume of labeled samples and this method is invariant to the scale transformation of the parameters. Using a simulation example and the application of two real data sets, the author demonstrates the success of these new methods in dealing with the label switching problem.     

Bayesian analysis of censored linear regression models with scale mixtures of normal distributions

As is the case of many studies, the data collected are limited and an exact value is recorded only if it falls within an interval range. Hence, the responses can be either left, interval or right censored. Linear (and nonlinear) regression models are routinely used to analyze these types of data and are based on normality assumptions for the errors terms. However, those analyzes might not provide robust inference when the normality assumptions are questionable. In this article, we develop a Bayesian framework for censored linear regression models by replacing the Gaussian assumptions for the random errors with scale mixtures of normal (SMN) distributions. The SMN is an attractive class of symmetric heavy-tailed densities that includes the normal, Student-t, Pearson type VII, slash and the contaminated normal distributions, as special cases. Using a Bayesian paradigm, an efficient Markov chain Monte Carlo algorithm is introduced to carry out posterior inference. A new hierarchical prior distribution is suggested for the degrees of freedom parameter in the Student-t distribution. The likelihood function is utilized to compute not only some Bayesian model selection measures but also to develop Bayesian case-deletion influence diagnostics based on the q-divergence measure. The proposed Bayesian methods are implemented in the R package BayesCR. The newly developed procedures are illustrated with applications using real and simulated data.