Intrinsic Priors for Model Selection Using an Encompassing Model with Applications to Censored Failure Time Data

In Bayesian model selection or testingproblems one cannot utilize standard or default noninformativepriors, since these priors are typically improper and are definedonly up to arbitrary constants. Therefore, Bayes factors andposterior probabilities are not well defined under these noninformativepriors, making Bayesian model selection and testing problemsimpossible. We derive the intrinsic Bayes factor (IBF) of Bergerand Pericchi (1996a, 1996b) for the commonly used models in reliabilityand survival analysis using an encompassing model. We also deriveproper intrinsic priors for these models, whose Bayes factors are asymptoticallyequivalent to the respective IBFs. We demonstrate our resultsin three examples.     

Objective Bayesian Testing of a Poisson Mean

We consider testing hypotheses about a single Poisson mean. When prior information is not available, use of objective priors is of interest. We provide intrinsic priors based on the arithmetic intrinsic and fractional Bayes factors, and evaluate their characteristics.     

Default Bayesian analysis of the Behrens–Fisher problem

In the Bayesian approach, the Behrens–Fisher problem has been posed as one of estimation for the difference of two means. No Bayesian solution to the Behrens–Fisher testing problem has yet been given due, perhaps, to the fact that the conventional priors used are improper. While default Bayesian analysis can be carried out for estimation purposes, it poses difficulties for testing problems. This paper generates sensible intrinsic and fractional prior distributions for the Behrens–Fisher testing problem from the improper priors commonly used for estimation. It allows us to compute the Bayes factor to compare the null and the alternative hypotheses. This default procedure of model selection is compared with a frequentist test and the Bayesian information criterion. We find discrepancy in the sense that frequentist and Bayesian information criterion reject the null hypothesis for data, that the Bayes factor for intrinsic or fractional priors do not.     

Alternative Bayes factors for model selection

Several alternative Bayes factors have been recently proposed in order to solve the problem of the extreme sensitivity of the Bayes factor to the priors of models under comparison. Specifically, the impossibility of using the Bayes factor with standard noninformative priors for model comparison has led to the introduction of new automatic criteria, such as the posterior Bayes factor (Aitkin 1991), the intrinsic Bayes factors (Berger and Pericchi 1996b) and the fractional Bayes factor (O'Hagan 1995). We derive some interesting properties of the fractional Bayes factor that provide justifications for its use additional to the ones given by O'Hagan. We further argue that the use of the fractional Bayes factor, originally introduced to cope with improper priors, is also useful in a robust analysis. Finally, using usual classes of priors, we compare several alternative Bayes factors for the problem of testing the point null hypothesis in the univariate normal model.     

Equality and inequality constrained multivariate linear models: Objective model selection using constrained posterior priors

In objective Bayesian model selection, a well-known problem is that standard non-informative prior distributions cannot be used to obtain a sensible outcome of the Bayes factor because these priors are improper. The use of a small part of the data, i.e., a training sample, to obtain a proper posterior prior distribution has become a popular method to resolve this issue and seems to result in reasonable outcomes of default Bayes factors, such as the intrinsic Bayes factor or a Bayes factor based on the empirical expected-posterior prior.     

INTRINSIC PRIORS FOR TWO-SAMPLE TESTS IN NORMAL POPULATIONS

ABSTRACT

There have been considerable amounts of work regarding the development of various default Bayes factors in model selection and hypothesis testing. Two commonly used criteria, the intrinsic Bayes factor and the fractional Bayes factor are compared to test two independent normal means and variances. We also derive several intrinsic priors whose Bayes factors are asymptotically equivalent to the respective Bayes factors. We demonstrate our results in simulated datasets.     

Restricted most powerful Bayesian tests for linear models

Uniformly most powerful Bayesian tests (UMPBTs) are a new class of Bayesian tests in which null hypotheses are rejected if their Bayes factor exceeds a specified threshold. The alternative hypotheses in UMPBTs are defined to maximize the probability that the null hypothesis is rejected. Here, we generalize the notion of UMPBTs by restricting the class of alternative hypotheses over which this maximization is performed, resulting in restricted most powerful Bayesian tests (RMPBTs). We then derive RMPBTs for linear models by restricting alternative hypotheses to g priors. For linear models, the rejection regions of RMPBTs coincide with those of usual frequentist F‐tests, provided that the evidence thresholds for the RMPBTs are appropriately matched to the size of the classical tests. This correspondence supplies default Bayes factors for many common tests of linear hypotheses. We illustrate the use of RMPBTs for ANOVA tests and t‐tests and compare their performance in numerical studies.     

Model determination for the variance component model using reference priors

For the balanced variance component model when the inference concerning intraclass correlation coefficient is of interest, Bayesian analysis is often appropriate. However, the question remains is to choose the appropriate prior. In this paper, we consider testing of the intraclass correlation coefficient under a default prior specification. Berger and Bernardo's (1992) On the development of the reference prior method. In: Bernardo, J.M., Berger, J.O., Dawid, A.P., Smith, A.F.M. (Eds.), Bayesian Statist. Vol. 4. Oxford University Press, London, pp. 35–60 reference priors are developed and are used to obtain the intrinsic Bayes factor (Berger and Pericchi, 1996) The intrinsic Bayes factor for model selection and prediction. J. Amer. statist. Assoc. 91, 109–122 for the nested models. Influence diagnostics using intrinsic Bayes factors are also developed. Finally, one simulated data is provided which illustrates the proposed methodology with appropriate simulation based on computational formulas. Then in order to overcome the difficulty in Bayesian computation, MCMC method, such as Gibbs sampler and Metropolis–Hastings algorithm, is employed.     

Objective priors for hypothesis testing in one-way random effects models

The Bayes factor is a key tool in hypothesis testing. Nevertheless, the important issue of which priors should be used to develop objective Bayes factors remains open. The authors consider this problem in the context of the one-way random effects model. They use concepts such as orthogonality, predictive matching and invariance to justify a specific form of the priors for common parameters and derive the intrinsic and divergence based prior for the new parameter. The authors show that both intrinsic priors or divergence-based priors produce consistent Bayes factors. They illustrate the methods and compare them with other proposals.     

Intrinsic Priors for Testing Two Normal Means with Intrinsic Bayes Factors

ABSTRACT

In this article we consider the problem of comparing two normal means with unknown common variance using a Bayesian approach. Conventional Bayes factors with improper non informative priors are not well defined. The intrinsic Bayes factors are used to overcome such a difficulty. We derive intrinsic priors whose Bayes factors are asymptotically equivalent to the corresponding intrinsic Bayes factors. We illustrate our results with numerical examples.     

Jeffreys priors for survival models with censored data

When prior information on model parameters is weak or lacking, Bayesian statistical analyses are typically performed with so-called “default” priors. We consider the problem of constructing default priors for the parameters of survival models in the presence of censoring, using Jeffreys’ rule. We compare these Jeffreys priors to the “uncensored” Jeffreys priors, obtained without considering censored observations, for the parameters of the exponential and log-normal models. The comparison is based on the frequentist coverage of the posterior Bayes intervals obtained from these prior distributions.     

Expansion estimation by Bayes rules

In the problem of estimating a location parameter in any symmetric unimodal location parameter model, we demonstrate that Bayes rules with respect to squared error loss can be expanders for some priors that belong to the family of all symmetric priors. That generalizes the results obtained by DasGupta and Rubin for the one dimensional case. We also consider symmetric priors which either have an appropriate point mass at 0 or are unimodal, and prove that under the latter condition all Bayes rules are shrinkers. Results of such nature are important, for example, in wavelet based function estimation and data denoising, where shrinkage of wavelet coefficients is associated with smoothing the data. We illustrate the results using FIAT stock market data.     

Posterior robustness with more than one sampling model

In order to robustify posterior inference, besides the use of large classes of priors, it is necessary to consider uncertainty about the sampling model. In this article we suggest that a convenient and simple way to incorporate model robustness is to consider a discrete set of competing sampling models, and combine it with a suitable large class of priors. This set reflects foreseeable departures of the base model, like thinner or heavier tails or asymmetry. We combine the models with different classes of priors that have been proposed in the vast literature on Bayesian robustness with respect to the prior. Also we explore links with the related literature of stable estimation and precise measurement theory, now with more than one model entertained. To these ends it will be necessary to introduce a procedure for model comparison that does not depend on an arbitrary constant or scale. We utilize a recent development on automatic Bayes factors with self-adjusted scale, the ‘intrinsic Bayes factor’ (Berger and Pericchi, Technical Report, 1993).     

Bayesian model comparison based on expected posterior priors for discrete decomposable graphical models

The implementation of the Bayesian paradigm to model comparison can be problematic. In particular, prior distributions on the parameter space of each candidate model require special care. While it is well known that improper priors cannot be routinely used for Bayesian model comparison, we claim that also the use of proper conventional priors under each model should be regarded as suspicious, especially when comparing models having different dimensions. The basic idea is that priors should not be assigned separately under each model; rather they should be related across models, in order to acquire some degree of compatibility, and thus allow fairer and more robust comparisons. In this connection, the intrinsic prior as well as the expected posterior prior (EPP) methodology represent a useful tool. In this paper we develop a procedure based on EPP to perform Bayesian model comparison for discrete undirected decomposable graphical models, although our method could be adapted to deal also with directed acyclic graph models. We present two possible approaches. One based on imaginary data, and one which makes use of a limited number of actual data. The methodology is illustrated through the analysis of a 2×3×4 contingency table.     

Bayesian model selection for life time data under type II censoring

This paper considers the Bayesian model selection problem in life-time models using type-II censored data. In particular, the intrinsic Bayes factors are calculated for log-normal, exponential, and Weibull lifetime models using noninformative priors under type-II censoring. Numerical examples are given to illustrate our results.     

Bayesian estimation and prediction based on lognormal record values

In this paper we consider the problems of estimation and prediction when observed data from a lognormal distribution are based on lower record values and lower record values with inter-record times. We compute maximum likelihood estimates and asymptotic confidence intervals for model parameters. We also obtain Bayes estimates and the highest posterior density (HPD) intervals using noninformative and informative priors under square error and LINEX loss functions. Furthermore, for the problem of Bayesian prediction under one-sample and two-sample framework, we obtain predictive estimates and the associated predictive equal-tail and HPD intervals. Finally for illustration purpose a real data set is analyzed and simulation study is conducted to compare the methods of estimation and prediction.     

Bayes admissible estimation of the means in Poisson decomposable graphical models

We investigate the Bayes estimation of the means in Poisson decomposable graphical models. Some classes of Bayes estimators are provided which improve on the maximum likelihood estimator under the normalized squared error loss. Both proper and improper priors are included in the proposed classes of priors. Concerning the generalized Bayes estimators with respect to the improper priors, we address their admissibility.     

Inference for Weibull distribution based on progressively Type-II hybrid censored data

Progressive Type-II hybrid censoring is a mixture of progressive Type-II and hybrid censoring schemes. In this paper, we discuss the statistical inference on Weibull parameters when the observed data are progressively Type-II hybrid censored. We derive the maximum likelihood estimators (MLEs) and the approximate maximum likelihood estimators (AMLEs) of the Weibull parameters. We then use the asymptotic distributions of the maximum likelihood estimators to construct approximate confidence intervals. Bayes estimates and the corresponding highest posterior density credible intervals of the unknown parameters are obtained under suitable priors on the unknown parameters and also by using the Gibbs sampling procedure. Monte Carlo simulations are then performed for comparing the confidence intervals based on all those different methods. Finally, one data set is analyzed for illustrative purposes.     

Bayes estimation and prediction of the two-parameter gamma distribution

In this article, the Bayes estimates of two-parameter gamma distribution are considered. It is well known that the Bayes estimators of the two-parameter gamma distribution do not have compact form. In this paper, it is assumed that the scale parameter has a gamma prior and the shape parameter has any log-concave prior, and they are independently distributed. Under the above priors, we use Gibbs sampling technique to generate samples from the posterior density function. Based on the generated samples, we can compute the Bayes estimates of the unknown parameters and can also construct HPD credible intervals. We also compute the approximate Bayes estimates using Lindley's approximation under the assumption of gamma priors of the shape parameter. Monte Carlo simulations are performed to compare the performances of the Bayes estimators with the classical estimators. One data analysis is performed for illustrative purposes. We further discuss the Bayesian prediction of future observation based on the observed sample and it is seen that the Gibbs sampling technique can be used quite effectively for estimating the posterior predictive density and also for constructing predictive intervals of the order statistics from the future sample.     

Radiocarbon dating with temporal order constraints

Bayesian methods are now widely used for analysing radiocarbon dates. We find that the non-informative priors in use in the literature generate a bias towards wider date ranges which does not in general reflect substantial prior knowledge. We recommend using a prior in which the distribution of the difference between the earliest and latest dates has a uniform distribution. We show how such priors are derived from a simple physical model of the deposition and observation process. We illustrate this in a case-study, examining the effect that various priors have on the reconstructed dates. Bayes factors are used to help to decide model choice problems.