How vague is vague? A simulation study of the impact of the use of vague prior distributions in MCMC using WinBUGS - PubMed (original) (raw)
Comparative Study
. 2005 Aug 15;24(15):2401-28.
doi: 10.1002/sim.2112.
Affiliations
- PMID: 16015676
- DOI: 10.1002/sim.2112
Comparative Study
How vague is vague? A simulation study of the impact of the use of vague prior distributions in MCMC using WinBUGS
Paul C Lambert et al. Stat Med. 2005.
Abstract
There has been a recent growth in the use of Bayesian methods in medical research. The main reasons for this are the development of computer intensive simulation based methods such as Markov chain Monte Carlo (MCMC), increases in computing power and the introduction of powerful software such as WinBUGS. This has enabled increasingly complex models to be fitted. The ability to fit these complex models has led to MCMC methods being used as a convenient tool by frequentists, who may have no desire to be fully Bayesian. Often researchers want 'the data to dominate' when there is no prior information and thus attempt to use vague prior distributions. However, with small amounts of data the use of vague priors can be problematic. The results are potentially sensitive to the choice of prior distribution. In general there are fewer problems with location parameters. The main problem is with scale parameters. With scale parameters, not only does one have to decide the distributional form of the prior distribution, but also whether to put the prior distribution on the variance, standard deviation or precision. We have conducted a simulation study comparing the effects of 13 different prior distributions for the scale parameter on simulated random effects meta-analysis data. We varied the number of studies (5, 10 and 30) and compared three different between-study variances to give nine different simulation scenarios. One thousand data sets were generated for each scenario and each data set was analysed using the 13 different prior distributions. The frequentist properties of bias and coverage were investigated for the between-study variance and the effect size. The choice of prior distribution was crucial when there were just five studies. There was a large variation in the estimates of the between-study variance for the 13 different prior distributions. With a large number of studies the choice of prior distribution was less important. The effect size estimated was not biased, but the precision with which it was estimated varied with the choice of prior distribution leading to varying coverage intervals and, potentially, to different statistical inferences. Again there was less of a problem with a larger number of studies. There is a particular problem if the between-study variance is close to the boundary at zero, as MCMC results tend to produce upwardly biased estimates of the between-study variance, particularly if inferences are based on the posterior mean. The choice of 'vague' prior distribution can lead to a marked variation in results, particularly in small studies. Sensitivity to the choice of prior distribution should always be assessed.
Copyright 2005 John Wiley & Sons, Ltd.
Similar articles
- Data cloning: easy maximum likelihood estimation for complex ecological models using Bayesian Markov chain Monte Carlo methods.
Lele SR, Dennis B, Lutscher F. Lele SR, et al. Ecol Lett. 2007 Jul;10(7):551-63. doi: 10.1111/j.1461-0248.2007.01047.x. Ecol Lett. 2007. PMID: 17542934 - Hierarchical Bayesian estimates of distributed MEG sources: theoretical aspects and comparison of variational and MCMC methods.
Nummenmaa A, Auranen T, Hämäläinen MS, Jääskeläinen IP, Lampinen J, Sams M, Vehtari A. Nummenmaa A, et al. Neuroimage. 2007 Apr 1;35(2):669-85. doi: 10.1016/j.neuroimage.2006.05.001. Epub 2007 Feb 12. Neuroimage. 2007. PMID: 17300961 - Multilevel modelling of clustered grouped survival data using Cox regression model: an application to ART dental restorations.
Wong MC, Lam KF, Lo EC. Wong MC, et al. Stat Med. 2006 Feb 15;25(3):447-57. doi: 10.1002/sim.2235. Stat Med. 2006. PMID: 16143989 - [Statistical models for spatial analysis in parasitology].
Biggeri A, Catelan D, Dreassi E, Lagazio C, Cringoli G. Biggeri A, et al. Parassitologia. 2004 Jun;46(1-2):75-8. Parassitologia. 2004. PMID: 15305691 Review. Italian. - Bayesian perspectives for epidemiological research. II. Regression analysis.
Greenland S. Greenland S. Int J Epidemiol. 2007 Feb;36(1):195-202. doi: 10.1093/ije/dyl289. Epub 2007 Feb 28. Int J Epidemiol. 2007. PMID: 17329317 Review.
Cited by
- A systematic review and Bayesian meta-analysis provide evidence for an effect of acute physical activity on cognition in young adults.
Garrett J, Chak C, Bullock T, Giesbrecht B. Garrett J, et al. Commun Psychol. 2024 Aug 28;2(1):82. doi: 10.1038/s44271-024-00124-2. Commun Psychol. 2024. PMID: 39242965 Free PMC article. - Practical parameter identifiability and handling of censored data with Bayesian inference in mathematical tumour models.
Porthiyas J, Nussey D, Beauchemin CAA, Warren DC, Quirouette C, Wilkie KP. Porthiyas J, et al. NPJ Syst Biol Appl. 2024 Aug 14;10(1):89. doi: 10.1038/s41540-024-00409-6. NPJ Syst Biol Appl. 2024. PMID: 39143084 Free PMC article. - ZIBGLMM: Zero-Inflated Bivariate Generalized Linear Mixed Model for Meta-Analysis with Double-Zero-Event Studies.
Li L, Lin L, Cappelleri JC, Chu H, Chen Y. Li L, et al. medRxiv [Preprint]. 2024 Jul 25:2024.07.25.24310959. doi: 10.1101/2024.07.25.24310959. medRxiv. 2024. PMID: 39108504 Free PMC article. Preprint. - Everything, altogether, all at once: Addressing data challenges when measuring speech intelligibility through entropy scores.
Espejo JMR, De Maeyer S, Gillis S. Espejo JMR, et al. Behav Res Methods. 2024 Oct;56(7):8132-8154. doi: 10.3758/s13428-024-02457-6. Epub 2024 Jul 24. Behav Res Methods. 2024. PMID: 39048860 Free PMC article. - RIMeta: An R shiny tool for estimating the reference interval from a meta-analysis.
Jiang Z, Cao W, Chu H, Bazerbachi F, Siegel L. Jiang Z, et al. Res Synth Methods. 2023 May;14(3):468-478. doi: 10.1002/jrsm.1626. Epub 2023 Feb 15. Res Synth Methods. 2023. PMID: 36725922 Free PMC article. Review.