Case Influence Analysis in Bayesian Inference

Abstract

We demonstrate how case influence analysis, commonly used in regression, can be applied to Bayesian hierarchical models. Draws from the joint posterior distribution of parameters are importance weighted to reflect the effect of deleting each observation in turn; the ensuing changes in the posterior distribution of each parameter are displayed graphically. The procedure is particularly useful when drawing a sample from the posterior distribution requires extensive calculations (as with a Markov Chain Monte Carlo sampler). The structure of hierarchical models, and other models with local dependence, makes the importance weights inexpensive to calculate with little additional programming. Some new alternative weighting schemes are described that extend the range of problems in which reweighting can be used to assess influence. Applications to a growth curve model and a complex hierarchical model for opinion data are described. Our focus on case influence on parameters is complementary to other work that measures influence by distances between posterior or predictive distributions.     

Reliability analysis for Weibull distribution with homogeneous heavily censored data based on Bayesian and least-squares methods

The reliability for Weibull distribution with homogeneous heavily censored data is analyzed in this study. The universal model of heavily censored data and existing methods, including maximum likelihood, least-squares, E-Bayesian estimation, and hierarchical Bayesian methods, are introduced. An improved method is proposed based on Bayesian inference and least-squares method. In this method, the Bayes estimations of failure probabilities are focused on for all the samples. The conjugate prior distribution of failure probability is set, and an optimization model is developed by maximizing the information entropy of prior distribution to determine the hyper-parameters. By integrating the likelihood function, the posterior distribution of failure probability is then derived to yield the Bayes estimation of failure probability. The estimations of reliability parameters are obtained by fitting distribution curve using least-squares method. The four existing methods are compared with the proposed method in terms of applicability, precision, efficiency, robustness, and simplicity. Specifically, the closed form expressions concerning E-Bayesian estimation and hierarchical Bayesian methods are derived and used. The comparisons demonstrate that the improved method is superior. Finally, three illustrative examples are presented to show the application of the proposed method.     

A prediction of individual growth of height according to an empirical Bayesian approach

An empirical Bayesian approach is applied to a prediction of an individual growth in height at an early stage of life. The sample has 548 normal growth of Japanese girls whose measurements are available on request. The prior distribution of estimator of the growth parameter vector in a lifetime growth model is obtained conventionally from the least squares estimates of the growth parameters. The choice of prior distributions is discussed from a practical point of view. It is possible to obtain a relevant prediction of growth based upon only measurements during the first six years of life. The lifetime prediction of individual growth at the age of 6 is enough approximation of real measurements obtained. This report deals with the comparison between the least squares estimates and an empirical Bayes estimates of the growth parameters and the characteristic points of the growth curve. We discuss the mean-constant growth curves of the groups classified by the height intervals at the age of 6.This work was supported in part by the ISM Cooperative Research Program (2-ISM·CRP-63).     

Prediction of soil water retention curve using Bayesian updating from limited measurement data

A soil water retention curve is one of the fundamental elements used to describe unsaturated soil. The accurate determination of soil water retention curve requires sufficient available information. However, the amount of measurement data is generally limited due to the restriction of time or test apparatus. As a result, it is a challenge to determine the soil water retention curve from limited measurement data. To address this problem, a Bayesian framework is proposed. In the Bayesian framework, Bayesian updating can be employed using the posterior distribution that is obtained by the Markov chain Monte Carlo sampling method with the Delayed Rejection Adaptive Metropolis algorithm. The parameters of soil water retention curve model are represented by the sample statistics of updating posterior distribution. A new updating algorithm based on Bayesian framework is proposed to predict the soil water retention curve using the ideal data and the limited measurement data of the granite residual soil and sand. The results show that the proposed prediction algorithm exhibits an excellent capability for more accurately determining the soil water retention curve with limited measured data. The uncertainty of updating parameters and the influence of the prior knowledge can be reduced. The converged results can be derived using the proposed prediction algorithm even if the prior knowledge is incomplete.     

BAYESIAN LOCAL INFLUENCE ASSESSMENTS IN A GROWTH CURVE MODEL WITH GENERAL COVARIANCE STRUCTURE

1 IntroductionThe problem of how to deal With local influence assessment in a growth curve model withgeneral covariance structure is very important. There are two main reasons why this is so. First,although the growth curve model can be viewed as a generalizetion of classical linear regressionmodel in some wad, as pointed out by for etc.[1], two models are substantially different andthe former is much more complicated than the latter. Secondly3 it is not generally the case withlocally influen…     

A Bayesian finite element model updating with combined normal and lognormal probability distributions using modal measurements

The present work is associated with Bayesian finite element (FE) model updating using modal measurements based on maximizing the posterior probability instead of any sampling based approach. Such Bayesian updating framework usually employs normal distribution in updating of parameters, although normal distribution has usual statistical issues while using non-negative parameters. These issues are proposed to be dealt with incorporating lognormal distribution for non-negative parameters. Detailed formulations are carried out for model updating, uncertainty-estimation and probabilistic detection of changes/damages of structural parameters using combined normal-lognormal probability distribution in this Bayesian framework. Normal and lognormal distributions are considered for eigen-system equation and structural (mass and stiffness) parameters respectively, while these two distributions are jointly considered for likelihood function. Important advantages in FE model updating (e.g. utilization of incomplete measured modal data, non-requirement of mode-matching) are also retained in this combined normal-lognormal distribution based proposed FE model updating approach. For demonstrating the efficiency of this proposed approach, a two dimensional truss structure is considered with multiple damage cases. Satisfactory performances are observed in model updating and subsequent probabilistic estimations, however level of performances are found to be weakened with increasing levels in damage scenario (as usual). Moreover, performances of this proposed FE model updating approach are compared with the typical normal distribution based updating approach for those damage cases demonstrating quite similar level of performances. The proposed approach also demonstrates better computational efficiency (achieving higher accuracy in lesser computation time) in comparison with two prominent Markov Chain Monte Carlo (MCMC) techniques (viz. Metropolis-Hastings algorithm and Gibbs sampling).     

增长曲线模型（一般协方差结构）的Bayes影响分析

本文讨论具有一般协方差结构的增长曲线模型中未知参数矩阵的Ｂａｙｅｓ影响分析问题．在无信息先验分布假设下,Ｋ－Ｌ距离被用来评估指定响应阵对参数矩阵的后验分布的影响程度．     

Classical and Bayesian inference in two parameter exponential distribution with randomly censored data

This paper deal with the classical and Bayesian estimation for two parameter exponential distribution having scale and location parameters with randomly censored data. The censoring time is also assumed to follow a two parameter exponential distribution with different scale but same location parameter. The main stress is on the location parameter in this paper. This parameter has not yet been studied with random censoring in literature. Fitting and using exponential distribution on the range \((0, \infty )\), specially when the minimum observation in the data set is significantly large, will give estimates far from accurate. First we obtain the maximum likelihood estimates of the unknown parameters with their variances and asymptotic confidence intervals. Some other classical methods of estimation such as method of moment, L-moments and least squares are also employed. Next, we discuss the Bayesian estimation of the unknown parameters using Gibbs sampling procedures under generalized entropy loss function with inverted gamma priors and Highest Posterior Density credible intervals. We also consider some reliability and experimental characteristics and their estimates. A Monte Carlo simulation study is performed to compare the proposed estimates. Two real data examples are given to illustrate the importance of the location parameter.     

Imprecise probability models for learning multinomial distributions from data. Applications to learning credal networks

This paper considers the problem of learning multinomial distributions from a sample of independent observations. The Bayesian approach usually assumes a prior Dirichlet distribution about the probabilities of the different possible values. However, there is no consensus on the parameters of this Dirichlet distribution. Here, it will be shown that this is not a simple problem, providing examples in which different selection criteria are reasonable. To solve it the Imprecise Dirichlet Model (IDM) was introduced. But this model has important drawbacks, as the problems associated to learning from indirect observations. As an alternative approach, the Imprecise Sample Size Dirichlet Model (ISSDM) is introduced and its properties are studied. The prior distribution over the parameters of a multinomial distribution is the basis to learn Bayesian networks using Bayesian scores. Here, we will show that the ISSDM can be used to learn imprecise Bayesian networks, also called credal networks when all the distributions share a common graphical structure. Some experiments are reported on the use of the ISSDM to learn the structure of a graphical model and to build supervised classifiers.     

Penalized marginal likelihood estimation of finite mixtures of Archimedean copulas

This paper proposes finite mixtures of different Archimedean copula families as a flexible tool for modelling the dependence structure in multivariate data. A novel approach to estimating the parameters in this mixture model is presented by maximizing the penalized marginal likelihood via iterative quadratic programming. The motivation for the penalized marginal likelihood stems from an underlying Bayesian model that imposes a prior distribution on the parameter of each Archimedean copula family. An approximative marginal likelihood is obtained by a classical quadrature discretization of the integral w.r.t. each family-specific prior distribution, thus yielding a finite mixture model. Family-specific smoothness penalties are added and the penalized marginal likelihood is maximized using an iterative quadratic programming routine. For comparison purposes, we also present a fully Bayesian approach via simulation-based posterior computation. The performance of the novel estimation approach is evaluated by simulations and two examples involving the modelling of the interdependence of exchange rates and of wind speed measurements, respectively. For these examples, penalized marginal likelihood estimates are compared to the corresponding Bayesian estimates.