Bayesian Model Selection for Longitudinal Count Data
نویسندگان
چکیده
We explore the performance of three popular model-selection criteria for generalised linear mixed-effects models (GLMMs) longitudinal count data (LCD). focus on evaluating conditional (given random effects) versus marginal (averaging over in selecting appropriate data-generating model. advocate use criteria, since Bayesian statisticians often despite previous warnings. discuss how to compute LCD by a replication method and importance sampling algorithm. Besides, we show via simulations what extent err when using instead criteria. To promote usage developed an R function that computes based samples from posterior distribution. Finally, illustrate advantages well-known set patients who have epilepsy.
منابع مشابه
Bayesian paradigm for analysing count data in longitudina studies using Poisson-generalized log-gamma model
In analyzing longitudinal data with counted responses, normal distribution is usually used for distribution of the random efffects. However, in some applications random effects may not be normally distributed. Misspecification of this distribution may cause reduction of efficiency of estimators. In this paper, a generalized log-gamma distribution is used for the random effects which includes th...
متن کاملA Bayesian model for longitudinal count data with non-ignorable dropout.
Asthma is an important chronic disease of childhood. An intervention programme for managing asthma was designed on principles of self-regulation and was evaluated by a randomized longitudinal study.The study focused on several outcomes, and, typically, missing data remained a pervasive problem. We develop a pattern-mixture model to evaluate the outcome of intervention on the number of hospitali...
متن کاملA State Space Model for Multivariate Longitudinal Count Data
A state space model for multivariate longitudinal count data driven by a latent gamma Markov process is proposed, the observed counts being conditionally independent and Poisson distributed given the latent process. We consider regression analysis for this model with time-varying covariates entering either via the Poisson model or via the latent gamma process. We develop the Kalman lter and smo...
متن کاملA serially correlated gamma frailty model for longitudinal count data
A Poisson-gamma model is introduced to account for between-subjects heterogeneity and within-subjects serial correlation occurring in longitudinal count data. The model extends the usual time-constant shared frailty approach to allow time-varying serially correlated gamma frailty whilst retaining standard marginal assumptions. A composite likelihood approach to estimation and testing for serial...
متن کاملA Bayesian moving average model for correlated count data
In this paper we propose a new regression modeling approach for responses that are a correlated time series of counts. The approach is based on a hierarchical Bayesian model that incorporates a latent moving average process with gamma random errors. This contrasts with the majority of existing models that remove correlation with an autoregressive process based on log-normal errors, and is likel...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Sankhya B
سال: 2021
ISSN: ['0976-8386', '0976-8394']
DOI: https://doi.org/10.1007/s13571-021-00268-9