Posterior Predictive Assessment of Model
نویسندگان
چکیده
This paper considers the Bayesian counterparts of the classical tests for goodness of t and their use in judging the t of a single Bayesian model to the observed data. We focus on posterior predictive assessment, in a framework that also includes conditioning on ancillary statistics. The Bayesian formulation facilitates the construction and calculation of a meaningful reference distribution not only for any (classical) statistic, but also for any parameter-dependent \statistic" or discrepancy variable. The latter allows us to propose the realized discrepancy assessment of model tness, which directly measures the true discrepancy between data and the posited model, for any aspect of the model which we want to explore. The computation required for the realized discrepancy assessment is a straightforward byproduct of the posterior simulation used for the original Bayesian analysis. We illustrate with three applied examples. The rst example, which serves mainly to motivate the work, illustrates the diiculty of classical tests in assessing the tness of a Poisson model to a positron emission tomography image that is constrained to be nonnegative. The second and third examples illustrate the details of the posterior predictive approach in two problems: estimation in a model with inequality constraints on the parameters, and estimation in a mixture model. In all three examples, standard test statistics (either a 2 or a likelihood ratio) are not pivotal: the diiculty is not just how to compute the reference distribution for the test, but that in the classical framework no such distribution exists, independent of the unknown model parameters.
منابع مشابه
Posterior Predictive Checks for Model Assessment in Occupancy Modeling
Though methodology and motivation for the use of posterior predictive checks has been introduced in some textbooks (e.g. Gelman et al. 2004), the technique is not commonplace in, for example, applied ecological studies using Bayesian techniques. In many studies using Bayesian techniques, the only semblance of model checking is the use of DIC for model selection, but this does not provide practi...
متن کاملPresenting a Model for Portfolio Risk Premium Assessment: Evidence from the Tehran Stock Exchange
This study aimed to present a model for portfolio risk premium assessment of companies listed in Tehran Stock Exchange. In order to achieve this purpose, monthly data of 150 companies listed in Tehran Stock Exchange during 2007-2017 was used. In this study, the predictive powers of FamaFrench three-factor model [11], Carhart four-factor model [1], Fama - French five-factor model [24], Brousseau...
متن کاملUse of posterior predictive assessments to evaluate model fit in multilevel logistic regression
Assessing the fit of a model is an important final step in any statistical analysis, but this is not straightforward when complex discrete response models are used. Cross validation and posterior predictions have been suggested as methods to aid model criticism. In this paper a comparison is made between four methods of model predictive assessment in the context of a three level logistic regres...
متن کاملPosterior Predictive Assessment of Model Fitness via Realized Discrepancies
This paper considers Bayesian counterparts of the classical tests for goodness of fit and their use in judging the fit of a single Bayesian model to the observed data. We focus on posterior predictive assessment, in a framework that also includes conditioning on auxiliary statistics. The Bayesian formulation facilitates the construction and calculation of a meaningful reference distribution not...
متن کاملThe Effects of the Predictive Ability of Accruals and Cash Flows on Earnings Quality: Evidences from TSE
This paper examines the effects of the predictive ability of accruals and cash flows on earnings quality in Tehran Stock Exchange (TSE). In this study we have used two methods: cross-section and pooled for testing hypothesis. The results show that Sloan model has the strongest ability to predict future earnings, and cash component earnings have more ability than accruals component earnings for ...
متن کامل