Reference Priors Research Articles

Objective Bayesian analysis for the accelerated degradation model using Wiener process with measurement errors

ABSTRACTThe Wiener process as a degradation model plays an important role in the degradation analysis. In this paper, we propose an objective Bayesian analysis for an acceleration degradation Wiener model which is subjected to measurement errors. The Jeffreys prior and reference priors under different group orderings are first derived, the propriety of the posteriors is then validated. It is shown that two of the reference priors can yield proper posteriors while the others cannot. A simulation study is carried out to investigate the frequentist performance of the approach compared to the maximum likelihood method. Finally, the approach is applied to analyse a real data.

The Prior Can Often Only Be Understood in the Context of the Likelihood

A key sticking point of Bayesian analysis is the choice of prior distribution, and there is a vast literature on potential defaults including uniform priors, Jeffreys’ priors, reference priors, maximum entropy priors, and weakly informative priors. These methods, however, often manifest a key conceptual tension in prior modeling: a model encoding true prior information should be chosen without reference to the model of the measurement process, but almost all common prior modeling techniques are implicitly motivated by a reference likelihood. In this paper we resolve this apparent paradox by placing the choice of prior into the context of the entire Bayesian analysis, from inference to prediction to model evaluation.

What should a forensic practitioner's likelihood ratio be? II

In the debate as to whether forensic practitioners should assess and report the precision of the strength of evidence statements that they report to the courts, I remain unconvinced by proponents of the position that only a subjectivist concept of probability is legitimate. I consider this position counterproductive for the goal of having forensic practitioners implement, and courts not only accept but demand, logically correct and scientifically valid evaluation of forensic evidence. In considering what would be the best approach for evaluating strength of evidence, I suggest that the desiderata be (1) to maximise empirically demonstrable performance; (2) to maximise objectivity in the sense of maximising transparency and replicability, and minimising the potential for cognitive bias; and (3) to constrain and make overt the forensic practitioner's subjective-judgement based decisions so that the appropriateness of those decisions can be debated before the judge in an admissibility hearing and/or before the trier of fact at trial. All approaches require the forensic practitioner to use subjective judgement, but constraining subjective judgement to decisions relating to selection of hypotheses, properties to measure, training and test data to use, and statistical modelling procedures to use – decisions which are remote from the output stage of the analysis – will substantially reduce the potential for cognitive bias. Adopting procedures based on relevant data, quantitative measurements, and statistical models, and directly reporting the output of the statistical models will also maximise transparency and replicability. A procedure which calculates a Bayes factor on the basis of relevant sample data and reference priors is no less objective than a frequentist calculation of a likelihood ratio on the same data. In general, a Bayes factor calculated using uninformative or reference priors will be closer to a value of 1 than a frequentist best estimate likelihood ratio. The bound closest to 1 based on a frequentist best estimate likelihood ratio and an assessment of its precision will also, by definition, be closer to a value of 1 than the frequentist best estimate likelihood ratio. From a practical perspective, both procedures shrink the strength of evidence value towards the neutral value of 1. A single-value Bayes factor or likelihood ratio may be easier for the courts to handle than a distribution. I therefore propose as a potential practical solution, the use of procedures which account for imprecision by shrinking the calculated Bayes factor or likelihood ratio towards 1, the choice of the particular procedure being based on empirical demonstration of performance.

Bayesian analysis of the generalized gamma distribution using non-informative priors

ABSTRACTThe Generalized gamma (GG) distribution plays an important role in statistical analysis. For this distribution, we derive non-informative priors using formal rules, such as Jeffreys prior, maximal data information prior and reference priors. We have shown that these most popular formal rules with natural ordering of parameters, lead to priors with improper posteriors. This problem is overcome by considering a prior averaging approach discussed in Berger et al. [Overall objective priors. Bayesian Analysis. 2015;10(1):189–221]. The obtained hybrid Jeffreys-reference prior is invariant under one-to-one transformations and yields a proper posterior distribution. We obtained good frequentist properties of the proposed prior using a detailed simulation study. Finally, an analysis of the maximum annual discharge of the river Rhine at Lobith is presented.

Objective Bayesian Entropy Inference for Two-Parameter Logistic Distribution Using Upper Record Values

In this paper, we provide an entropy inference method that is based on an objective Bayesian approach for upper record values having a two-parameter logistic distribution. We derive the entropy that is based on the i-th upper record value and the joint entropy that is based on the upper record values. Moreover, we examine their properties. For objective Bayesian analysis, we obtain objective priors, namely, the Jeffreys and reference priors, for the unknown parameters of the logistic distribution. The priors are based on upper record values. Then, we develop an entropy inference method that is based on these objective priors. In real data analysis, we assess the quality of the proposed models under the objective priors and compare them with the model under the informative prior.

Bayesian inference on extreme value distribution using upper record values

ABSTRACTIn this paper we address estimation and prediction problems for extreme value distributions under the assumption that the only available data are the record values. We provide some properties and pivotal quantities, and derive unbiased estimators for the location and rate parameters based on these properties and pivotal quantities. In addition, we discuss mean-squared errors of the proposed estimators and exact confidence intervals for the rate parameter. In Bayesian inference, we develop objective Bayesian analysis by deriving non informative priors such as the Jeffrey, reference, and probability matching priors for the location and rate parameters. We examine the validity of the proposed methods through Monte Carlo simulations for various record values of size and present a real data set for illustration purposes.

On the Flatland paradox and limiting arguments

ABSTRACTWe revisit the Flatland paradox proposed by Stone (1976), which is an example of non conglomerability. The main novelty in the analysis of the paradox is to consider marginal versus conditional models rather than proper versus improper priors. We show that in the first model a prior distribution should be considered as a probability measure, whereas, in the second one, a prior distribution should be considered in the projective space of measures. This induces two different kinds of limiting arguments which are useful to understand the paradox. We also show that the choice of a flat prior is not adapted to the structure of the parameter space and we consider an improper prior based on reference priors with nuisance parameters for which the Bayesian analysis matches the intuitive reasoning.

Bayesian analysis of repairable systems with modulated power law process

As a compromise between nonhomogeneous Poisson process and renewal process, the modulated power law process is more appropriate to model the failures of repairable systems. In this article, objective Bayesian methods are proposed to analyze the modulated power law process. Seven reference priors, one of which is also the Jeffreys prior, are derived. However, only four of them are taken into consideration because of their practicality. Propriety of the posterior densities considering the four reference priors is proved. Predictive distribution of the future failure time is obtained additionally. For the purpose of comparison, the simulation work and real data analysis are carried out based on both the objective Bayesian and maximum likelihood approaches, which show that the objective Bayesian estimation and prediction have much better statistical properties in a frequentist context, and outperforms the maximum likelihood method even with small or moderate sample sizes.

How Principled and Practical Are Penalised Complexity Priors?

This note discusses the paper Penalising model component by Simpson et al. (2017). While we acknowledge the highly novel approach to prior construction and commend the authors for setting new-encompassing principles that will Bayesian modelling, and while we perceive the potential connection with other branches of the literature, we remain uncertain as to what extent the principles exposed in the paper can be developed outside specific models, given their lack of precision. The very notions of model component, base model, overfitting prior are for instance conceptual rather than mathematical and we thus fear the concept of penalised complexity may not further than extending first-guess priors into larger families, thus failing to establish reference priors on a novel sound ground.

Bayesian analysis of series system with dependent causes of failure

ABSTRACTMost studies of series system assume the causes of failure are independent, which may not hold in practice. In this paper, dependent causes of failure are considered by using a Marshall–Olkin bivariate Weibull distribution. We derived four reference priors based on several grouping orders. Gibbs sampling combined with the rejection sampling algorithm and Metropolis–Hastings algorithm is developed to obtain the estimates of the unknown parameters. The proposed approach is compared with the maximum-likelihood method via simulation. We find that the root-mean-squared errors of the Bayesian estimates are much smaller for the case of small sample size, and that the coverage probabilities of the Bayesian estimates are much closer to the nominal levels. Finally, a real data-set is analysed for illustration.

Objective Bayesian analysis of JM model in software reliability

Jelinski Moranda (JM) model is frequently used in software reliability. The objective Bayesian inference was proposed to estimate the parameters of JM model. Jeffreys prior and reference priors have been derived. Besides, the properties of corresponding posteriors were deduced and some modifications were made which made the posterior distributions proper. Then Gibbs sampling was utilized to obtain the Bayesian estimators, credible intervals and coverage probabilities of the parameters. Comparisons in the efficiency of the maximum likelihood estimators and Bayesian estimators under different priors for various sample sizes have been done by simulations and a real data set was analyzed for illustrative purpose.

Bayesian estimation of the generalized lognormal distribution using objective priors

ABSTRACTThe generalized lognormal distribution plays an important role in analysing data from different life testing experiments. In this paper, we consider Bayesian analysis of this distribution using various objective priors for the model parameters. Specifically, we derive expressions for the Jeffreys-type priors, the reference priors with different group orderings of the parameters, and the first-order matching priors. We also study the properties of the posterior distributions of the parameters under these improper priors. It is shown that only two of them result in proper posterior distributions. Numerical simulation studies are conducted to compare the performances of the Bayesian estimators under the considered priors and the maximum likelihood estimates. Finally, a real-data application is also provided for illustrative purposes.

Consensus priors for multinomial and binomial ratios

ABSTRACTReference analysis is an objective Bayesian approach to finding noninformative prior distributions. For various models involving nuisance parameters the reference prior has been shown to be superior to the multiparameter Jeffreys prior. In this article, the performance of the reference prior is evaluated for models that are defined at the extremes of parameter ranges, in which case reference and Jeffreys priors are shown to be potentially informative. Two quantities of interest are analyzed: the ratio of two multinomial parameters and the ratio of two independent binomial parameters, where the latter is just one example of inference based on the common 2 × 2 contingency table. For these two specific examples, we show that reference and/or Jeffreys priors lead to overinformative inference in extreme data situations that would seem common in, for example, many medical contexts when interest is in rare events. It is recommended that in the context of noninformative priors and binary data with extreme...

A Bayesian control chart for a common coefficient of variation

ABSTRACTBy using the medical data analyzed by Kang et al. (2007), a Bayesian procedure is applied to obtain control limits for the coefficient of variation. Reference and probability matching priors are derived for a common coefficient of variation across the range of sample values. By simulating the posterior predictive density function of a future coefficient of variation, it is shown that the control limits are effectively identical to those obtained by Kang et al. (2007) for the specific dataset they used. This article illustrates the flexibility and unique features of the Bayesian simulation method for obtaining posterior distributions, predictive intervals, and run-lengths in the case of the coefficient of variation. A simulation study shows that the 95% Bayesian confidence intervals for the coefficient of variation have the correct frequentist coverage.

Bayesian testing for the homogeneity of the shape parameters of several inverse Gaussian distributions

We develop the testing procedures about the homogeneity of the shape parameters of several inverse Gaussian distributions in our paper. We propose default Bayesian testing procedures for the shape parameters under the reference priors. The Bayes factor based on the proper priors gives the successful results for Bayesian hypothesis testing. For the case of the lack of information, the noninformative priors such as Jereys' prior or the reference prior can be used. Jereys' prior or the reference prior involves the undefined constants in the computation of the Bayes factors. Therefore under the reference priors, we develop the Bayesian testing procedures with the intrinsic Bayes factors and the fractional Bayes factor. Simulation study for the performance of the developed testing procedures is given, and an example for illustration is given.

Objective Bayesian inference for the ratio of the scale parameters of two Weibull distributions

ABSTRACTThe Weibull distribution is widely used due to its versatility and relative simplicity. In our paper, the non informative priors for the ratio of the scale parameters of two Weibull models are provided. The asymptotic matching of coverage probabilities of Bayesian credible intervals is considered, with the corresponding frequentist coverage probabilities. We developed the various priors for the ratio of two scale parameters using the following matching criteria: quantile matching, matching of distribution function, highest posterior density matching, and inversion of test statistics. One particular prior, which meets all the matching criteria, is found. Next, we derive the reference priors for groups of ordering. We see that all the reference priors satisfy a first-order matching criterion and that the one-at-a-time reference prior is a second-order matching prior. A simulation study is performed and an example given.

Objective priors for the zero-modified model

We consider zero-inflated models, and use the Jeffreys, reference, and matching priors as objective prior distributions derived from the hurdle and with zeros models from the Bayesian viewpoint. We investigate the properties of the resulting posterior distributions.

An objective Bayesian analysis of the two-parameter half-logistic distribution based on progressively type-II censored samples

ABSTRACTThis paper develops an objective Bayesian analysis method for estimating unknown parameters of the half-logistic distribution when a sample is available from the progressively Type-II censoring scheme. Noninformative priors such as Jeffreys and reference priors are derived. In addition, derived priors are checked to determine whether they satisfy probability-matching criteria. The Metropolis–Hasting algorithm is applied to generate Markov chain Monte Carlo samples from these posterior density functions because marginal posterior density functions of each parameter cannot be expressed in an explicit form. Monte Carlo simulations are conducted to investigate frequentist properties of estimated models under noninformative priors. For illustration purposes, a real data set is presented, and the quality of models under noninformative priors is evaluated through posterior predictive checking.

Statistical inference on Gumbel distribution using record values

In this study, we address inference problems for Gumbel distribution when the available data are lower record values. We first derive unbiased estimators of unknown parameters, and then, we construct an exact confidence interval for the scale parameter and a predictive interval for the next lower value by deriving certain properties and pivotal quantities. These are compared with the results for existing inference. For Bayesian inference, we derive noninformative priors such as the Jeffreys and reference priors for unknown parameters and examine whether they satisfy the probability matching criteria; then, we apply them to develop objective Bayesian analysis.

A comparative simulation study of AR(1) estimators in short time series

Various estimators of the autoregressive model exist. We compare their performance in estimating the autocorrelation in short time series. In Study 1, under correct model specification, we compare the frequentist r1 estimator, C-statistic, ordinary least squares estimator (OLS) and maximum likelihood estimator (MLE), and a Bayesian method, considering flat (Bf) and symmetrized reference (Bsr) priors. In a completely crossed experimental design we vary lengths of time series (i.e., T = 10, 25, 40, 50 and 100) and autocorrelation (from −0.90 to 0.90 with steps of 0.10). The results show a lowest bias for the Bsr, and a lowest variability for r1. The power in different conditions is highest for Bsr and OLS. For T = 10, the absolute performance of all measurements is poor, as expected. In Study 2, we study robustness of the methods through misspecification by generating the data according to an ARMA(1,1) model, but still analysing the data with an AR(1) model. We use the two methods with the lowest bias for this study, i.e., Bsr and MLE. The bias gets larger when the non-modelled moving average parameter becomes larger. Both the variability and power show dependency on the non-modelled parameter. The differences between the two estimation methods are negligible for all measurements.