Improper Prior Distributions Research Articles

Bayesian estimation and prediction under progressive-stress accelerated life test for a log-logistic model

Accelerated life tests play a very critical role in reliability analysis because highly reliable products are being produced with recent advanced technologies to sustain the market demand and competition. A progressive-stress accelerated life test is one of the kinds of accelerated life tests that allows applied stress to change continuously. Considering the importance of progressive-stress accelerated life tests, this paper deals with progressive-stress accelerated life tests when the lifetimes of units follow the log-logistic distribution and are progressively type-II censored where the associated scale parameter conforms to the inverse power law. Different estimators of model parameters are derived using maximum likelihood and Bayesian methods. Interval estimation is also considered. In the sequel, approximate confidence, bootstrap, and Bayes credible intervals are constructed. Bayes estimators are obtained under squared error, LINEX, and entropy loss functions using proper and improper prior distributions. A simulation study is conducted based on various censoring schemes. The coverage percentages and interval widths are computed via Monte Carlo simulations. Bayes predictive estimates and intervals are also obtained. Finally, two different accelerated life test data are analyzed for illustration purposes.

Recurrence and transience of a Markov chain on $\mathbb Z$ + and evaluation of prior distributions for a Poisson mean

AbstractEaton (1992) considered a general parametric statistical model paired with an improper prior distribution for the parameter and proved that if a certain Markov chain, constructed using the model and the prior, is recurrent, then the improper prior is strongly admissible, which (roughly speaking) means that the generalized Bayes estimators derived from the corresponding posterior distribution are admissible. Hobert and Robert (1999) proved that Eaton’s Markov chain is recurrent if and only if its so-called conjugate Markov chain is recurrent. The focus of this paper is a family of Markov chains that contains all of the conjugate chains that arise in the context of a Poisson model paired with an arbitrary improper prior for the mean parameter. Sufficient conditions for recurrence and transience are developed and these are used to establish new results concerning the strong admissibility of non-conjugate improper priors for the Poisson mean.

Bayesian estimation and prediction for network meta-analysis with contrast-based approach.

Network meta-analysis is gaining prominence in clinical epidemiology and health technology assessments that enable comprehensive assessment of comparative effectiveness for multiple available treatments. In network meta-analysis, Bayesian methods have been one of the standard approaches for the arm-based approach and are widely applied in practical data analyses. Also, for most cases in these applications, proper noninformative priors are adopted, which does not incorporate subjective prior knowledge into the analyses, and reference Bayesian analyses are major choices. In this article, we provide generic Bayesian analysis methods for the contrast-based approach of network meta-analysis, where the generic Bayesian methods can treat proper and improper prior distributions. The proposed methods enable direct sampling from the posterior and posterior predictive distributions without formal iterative computations such as Markov chain Monte Carlo, and technical convergence checks are not required. In addition, representative noninformative priors that can be treated in the proposed framework involving the Jeffreys prior are provided. We also provide an easy-to-handle R statistical package, BANMA, to implement these Bayesian analyses via simple commands. The proposed Bayesian methods are illustrated using various noninformative priors through applications to two real network meta-analyses.

Estimation of P(Y

ABSTRACT The problem of estimating a stress-strength parameter is considered under the assumption that both stress and strength variables follow independent lognormal distributions. The maximum likelihood estimator of this parametric function is obtained and based on its asymptotic distribution, approximate intervals are constructed. Properties of a heuristic estimator are studied as well. Bootstrap intervals are also constructed. Bayes estimator is derived against proper and improper prior distributions. We further compute credible intervals of stress–strength parameter using importance sampling. The empirical Bayes procedure is also discussed. Proposed point and interval estimators are compared numerically using Monte Carlo simulations. Analysis of a real data set is presented for illustration purposes.

Posterior-based Wald-type statistics for hypothesis testing

A new Wald-type statistic is proposed for hypothesis testing based on Bayesian posterior distributions under the correct model specification. The new statistic can be explained as a posterior version of the Wald statistic and has several nice properties. First, it is well-defined under improper prior distributions. Second, it avoids Jeffreys–Lindley–Bartlett’s paradox. Third, under the null hypothesis and repeated sampling, it follows a χ2 distribution asymptotically, offering an asymptotically pivotal test. Fourth, it only requires inverting the posterior covariance for parameters of interest. Fifth and perhaps most importantly, when a random sample from the posterior distribution (such as MCMC output) is available, the proposed statistic can be easily obtained as a by-product of posterior simulation. In addition, the numerical standard error of the estimated proposed statistic can be computed based on random samples. A robust version of the test statistic is developed under model misspecification and inherits many nice properties of the new posterior statistic. The finite sample performance of the statistics is examined in Monte Carlo studies. The method is applied to two latent variable models used in microeconometrics and financial econometrics.

Simple and flexible Bayesian inferences for standardized regression coefficients

ABSTRACTIn statistical practice, inferences on standardized regression coefficients are often required, but complicated by the fact that they are nonlinear functions of the parameters, and thus standard textbook results are simply wrong. Within the frequentist domain, asymptotic delta methods can be used to construct confidence intervals of the standardized coefficients with proper coverage probabilities. Alternatively, Bayesian methods solve similar and other inferential problems by simulating data from the posterior distribution of the coefficients. In this paper, we present Bayesian procedures that provide comprehensive solutions for inferences on the standardized coefficients. Simple computing algorithms are developed to generate posterior samples with no autocorrelation and based on both noninformative improper and informative proper prior distributions. Simulation studies show that Bayesian credible intervals constructed by our approaches have comparable and even better statistical properties than their frequentist counterparts, particularly in the presence of collinearity. In addition, our approaches solve some meaningful inferential problems that are difficult if not impossible from the frequentist standpoint, including identifying joint rankings of multiple standardized coefficients and making optimal decisions concerning their sizes and comparisons. We illustrate applications of our approaches through examples and make sample R functions available for implementing our proposed methods.

How proper are Bayesian models in the astronomical literature?

The well-known Bayes theorem assumes that a posterior distribution is a probability distribution. However, the posterior distribution may no longer be a probability distribution if an improper prior distribution (non-probability measure) such as an unbounded uniform prior is used. Improper priors are often used in the astronomical literature to reflect a lack of prior knowledge, but checking whether the resulting posterior is a probability distribution is sometimes neglected. It turns out that 23 articles out of 75 articles (30.7%) published online in two renowned astronomy journals (ApJ and MNRAS) between Jan 1, 2017 and Oct 15, 2017 make use of Bayesian analyses without rigorously establishing posterior propriety. A disturbing aspect is that a Gibbs-type Markov chain Monte Carlo (MCMC) method can produce a seemingly reasonable posterior sample even when the posterior is not a probability distribution (Hobert and Casella, 1996). In such cases, researchers may erroneously make probabilistic inferences without noticing that the MCMC sample is from a non-existing probability distribution. We review why checking posterior propriety is fundamental in Bayesian analyses, and discuss how to set up scientifically motivated proper priors.

Approximate large-scale Bayesian spatial modeling with application to quantitative magnetic resonance imaging

We consider the Bayesian inference of nonlinear, large-scale regression problems in which the parameters model the spatial distribution of some property. A homoscedastic Gaussian sampling distribution is supposed as well as certain assumptions about the regression function. Propriety of the posterior and the existence of its moments are explored when using improper prior distributions expressing different levels of prior knowledge, ranging from a purely noninformative prior over intrinsic Gaussian Markov random field priors to a partition prior. The considered class of problems includes magnetic resonance fingerprinting (MRF). We apply an approximate Bayesian inference to this particular application and demonstrate its practicability in dimensions up to $$10^5$$ or larger. The benefit of incorporating substantial prior knowledge is illustrated. By analyzing simulated realistic MRF data, it is shown that MAP estimates can significantly improve the results achieved with maximum likelihood estimation.

Classical-Equivalent Bayesian Portfolio Optimization for Electricity Generation Planning.

There are several electricity generation technologies based on different sources such as wind, biomass, gas, coal, and so on. The consideration of the uncertainties associated with the future costs of such technologies is crucial for planning purposes. In the literature, the allocation of resources in the available technologies has been solved as a mean-variance optimization problem assuming knowledge of the expected values and the covariance matrix of the costs. However, in practice, they are not exactly known parameters. Consequently, the obtained optimal allocations from the mean-variance optimization are not robust to possible estimation errors of such parameters. Additionally, it is usual to have electricity generation technology specialists participating in the planning processes and, obviously, the consideration of useful prior information based on their previous experience is of utmost importance. The Bayesian models consider not only the uncertainty in the parameters, but also the prior information from the specialists. In this paper, we introduce the classical-equivalent Bayesian mean-variance optimization to solve the electricity generation planning problem using both improper and proper prior distributions for the parameters. In order to illustrate our approach, we present an application comparing the classical-equivalent Bayesian with the naive mean-variance optimal portfolios.

A Posterior-Based Wald-Type Statistic for Hypothesis Testing

A new Wald-type statistic is proposed for hypothesis testing based on Bayesian posterior distributions. The new statistic can be explained as a posterior version of Wald test and have several nice properties. First, it is well-defined under improper prior distributions. Second, it avoids Jeffreys-Lindley’s paradox. Third, under the null hypothesis and repeated sampling, it follows a χ2 distribution asymptotically, offering an asymptotically pivotal test. Fourth, it only requires inverting the posterior covariance for the parameters of interest. Fifth and perhaps most importantly, when a random sample from the posterior distribution (such as an MCMC output) is available, the proposed statistic can be easily obtained as a by-product of posterior simulation. In addition, the numerical standard error of the estimated proposed statistic can be computed based on the random sample. The finite sample performance of the statistic is examined in Monte Carlo studies. The method is applied to two latent variable models used in micro-econometrics and financial econometrics.

Frequency coverage properties of a uniform shrinkage prior distribution

ABSTRACTA uniform shrinkage prior (USP) distribution on the unknown variance component of a random-effects model is known to produce good frequency properties. The USP has a parameter that determines the shape of its density function, but it has been neglected whether the USP can maintain such good frequency properties regardless of the choice for the shape parameter. We investigate which choice for the shape parameter of the USP produces Bayesian interval estimates of random effects that meet their nominal confidence levels better than several existent choices in the literature. Using univariate and multivariate Gaussian hierarchical models, we show that the USP can achieve its best frequency properties when its shape parameter makes the USP behave similarly to an improper flat prior distribution on the unknown variance component.

Bayesian Estimation for the Centered Parameterization of the Skew-Normal Distribution

The skew-normal (SN) distribution is a generalization of the normal distribution, where a shape parameter is added to adopt skewed forms. The SN distribution has some of the properties of a univariate normal distribution, which makes it very attractive from a practical standpoint; however, it presents some inference problems. Specifically, the maximum likelihood estimator for the shape parameter tends to infinity with a positive probability. A new Bayesian approach is proposed in this paper which allows to draw inferences on the parameters of this distribution by using improper prior distributions in the ``centered parametrization'' for the location and scale parameter and a Beta-type for the shape parameter. Samples from posterior distributions are obtained by using the Metropolis-Hastings algorithm. A simulation study shows that the mode of the posterior distribution appears to be a good estimator in terms of bias and mean squared error. A comparative study with similar proposals for the SN estimation problem was undertaken. Simulation results provide evidence that the proposed method is easier to implement than previous ones. Some applications and comparisons are also included.

Mutational pattern of a sample from a critical branching population.

We study a universal object for the genealogy of a sample in populations with mutations: the critical birth-death process with Poissonian mutations, conditioned on its population size at a fixed time horizon. We show how this process arises as the law of the genealogy of a sample in a large class of nearly critical branching populations with rare mutations at birth, namely populations converging, in a large population asymptotic, towards the continuum random tree. We extend this model to populations with random foundation times, with (potentially improper) prior distributions [Formula: see text], [Formula: see text], including the so-called uniform ([Formula: see text]) and log-uniform ([Formula: see text]) priors. We first investigate the mutational patterns arising from these models, by studying the site frequency spectrum of a sample with fixed size, i.e. the number of mutations carried by k individuals in the sample. Explicit formulae for the expected frequency spectrum of a sample are provided, in the cases of a fixed foundation time, and of a uniform and log-uniform prior on the foundation time. Second, we establish the convergence in distribution, for large sample sizes, of the (suitably renormalized) tree spanned by the sample with prior [Formula: see text] on the time of origin. We finally prove that the limiting genealogies with different priors can all be embedded in the same realization of a given Poisson point measure.

Classification into Two-Parameter Exponential Populations with a Common Guarantee Time

SYNOPTIC ABSTRACTLet Π1, Π2, …, Πk be k (≥2) exponential populations with unknown scale parameters σ1, σ2, …, σk, respectively, and an unknown but common location parameter μ. First, we consider the estimation of scale parameters when an isotonic ordering among the scale parameters is present. We show the superiority of a class of mixed estimators over the maximum likelihood estimators of scale parameters under a scale-invariant loss function. Bayes and generalized Bayes estimates of scale parameters are obtained assuming proper and improper prior distributions, respectively. As an application of these new estimators, we have considered the problem of classifying an observation into one of k populations under order restrictions on scale parameters. Classification rules are proposed based on mixed estimators. We also derive plug-in Bayes classification rules and likelihood ratio-based classification rules. Extensive simulations are performed to compare these rules with respect to the expected probability of correct classification. An application of the classification rules is done on a real data set.

A Bayesian chi-squared test for hypothesis testing

A new Bayesian test statistic is proposed to test a point null hypothesis based on a quadratic loss. The proposed test statistic may be regarded as the Bayesian version of the Lagrange multiplier test. Its asymptotic distribution is obtained based on a set of regular conditions and follows a chi-squared distribution when the null hypothesis is correct. The new statistic has several important advantages that make it appealing in practical applications. First, it is well-defined under improper prior distributions. Second, it avoids Jeffrey–Lindley’s paradox. Third, it always takes a non-negative value and is relatively easy to compute, even for models with latent variables. Fourth, its numerical standard error is relatively easy to obtain. Finally, it is asymptotically pivotal and its threshold values can be obtained from the chi-squared distribution. The method is illustrated using some real examples in economics and finance.

Bayesian path specific frailty models for multi-state survival data with applications.

Multi-state models can be viewed as generalizations of both the standard and competing risks models for survival data. Models for multi-state data have been the theme of many recent published works. Motivated by bone marrow transplant data, we propose a Bayesian model using the gap times between two successive events in a path of events experienced by a subject. Path specific frailties are introduced to capture the dependence structure of the gap times in the paths with two or more states. Under improper prior distributions for the parameters, we establish propriety of the posterior distribution. An efficient Gibbs sampling algorithm is developed for drawing samples from the posterior distribution. An extensive simulation study is carried out to examine the empirical performance of the proposed approach. A bone marrow transplant data set is analyzed in detail to further demonstrate the proposed methodology.

Empirical Bayes Approach for Developing Hierarchical Probabilistic Predictive Models and Its Application to the Seismic Reliability Analysis of FRP-Retrofitted RC Bridges

AbstractThis paper proposes a general formulation to develop hierarchical probabilistic predictive models for clustered data. The common clustering factor, shared among the data within a group, causes statistical dependence that needs to be accounted for in the estimation of unknown model parameters. The basic idea of the hierarchical formulation is that the unknown model parameters are endowed with distributions that depend on a set of shared underlying parameters, and this construction is recursive up to the highest level of the hierarchy. The usual improper noninformative prior distributions on variance parameters of hierarchical models can lead to nonexistent posterior distributions that may appear perfectly reasonable in numerical simulations. On the other hand, common proper noninformative prior distributions may also substantially affect posterior statistics. Instead, the empirical Bayes approach is proposed to objectively estimate the variance parameters. The Gibbs sampling algorithm is used to es...

Bayesian analysis based on the Jeffreys prior for the hyperbolic distribution

In this work, we develop Bayesian analysis based on the Jeffreys prior for the hyperbolic family of distributions. It is usually difficult to estimate the four parameters in this class: to be reliable the maximum likelihood estimator typically requires large sample sizes of the order of thousands of observations. Moreover, improper prior distributions may lead to improper posterior distributions, whereas proper prior distributions may dominate the analysis. Here, we show through a simulation study that Bayesian methods based on Jeffreys prior provide reliable point and interval estimators. Moreover, this simulation study shows that for the absolute loss function Bayesian estimators compare favorably to maximum likelihood estimators. Finally, we illustrate with an application to real data that our methodology allows for parameter estimation with remarkable good properties even for a small sample size.

A stochastic simulation method for uncertainty quantification in the linearized inverse conductivity problem

SUMMARY This paper considers the inverse problem in electrical impedance tomography with non-informative prior information on the required conductivity function. The problem is approached with a Newton-type iterative algorithm where the solution of the linearized approximation is estimated using Bayesian inference. The novelty of this work focuses on maximum a posteriori estimation assuming a model that incorporates the linearization error as a random variable. From an analytical expression of this term, we employ Monte Carlo simulation in order to characterize its probability distribution function. This simulation entails sampling an improper prior distribution for which we propose a stable scheme on the basis of QR decomposition. The simulation statistics show that the error on the linearized model is not Gaussian, however, to maintain computational tractability, we derive the posterior probability density function of the solution by imposing a Gaussian kernel approximation to the error density. Numerical results obtained through this approach indicate the superiority of the new model and its respective maximum a posteriori estimator against the conventional one that neglects the impact of the linearization error. Copyright © 2011 John Wiley & Sons, Ltd.

Bayesian estimation of the parameter of a generalized geometric series distribution under different prior distributions and loss functions

Comparisons of estimates between the Bayes and frequentist methods are fascinating and challenging theme of interest in statistics with significant impact on professionals. For Bayes estimators, the performance depends on the form of the prior distribution and the assumed loss function. This paper illu strates the problem of estimation of the one-parameter generalized geometric series distribution, using conjugate and improper prior distributions under symmetric and asymmetric loss functions. Performance of the Bayes estimates with respect to different priors, loss functions, and maximum likelihood estimates are illustrated for a data set and through a simulation study.