How vague is vague? A simulation study of the impact of the use of vague prior distributions in MCMC using WinBUGS

By Paul C. Lambert, A. Sutton, P. Burton, K. Abrams, David R. Jones. 2005

533 citations recorded by the source corpus.

Subfields: Management Science and Operations Research, Statistics and Probability

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Management Science and Operations Research

In Statistics and Probability

Citations in the corpus

A Bayesian model for repeated measures zero-inflated count data with application to outpatient psychiatric service use (2010, background): “Some Statistical Modelling 2010; 10(4): 421–439 authors have noted that I-W priors can yield biased and imprecise model estimates, particularly if the number of clustering units is small (Browne and Draper, 2000; Lambert et al., 2005).”
Multivariate meta-analysis: Potential and promise (2011, background): “A recent investigation shows that alternative and apparently vague priors can produce markedly different results in the univariate setting [60] and this situation is likely to worsen as the dimension of the analysis, and hence the number of parameters, increases [61].”
Propriety of posteriors in structured additive regression models: Theory and empirical evidence (2009, background): “Although these (ε, ε)-priors lead to proper posteriors under the simplified conditions of Corollary 1, there is some debate in the literature about sensitivity of posteriors and numerical stability in practical work, see for example Lambert et al. (2005) and Gelman (2006).”
WinBUGS for population ecologists: bayesian modeling using markov chain Monte Carlo methods (2009, background): “Several authors have dealt with important issues regarding the specification of vague priors (Lambert et al. 2005; van Dongen 2006; Gelman 2006), assessment of the sensitivity of the posterior distribution to the specified prior distribution (Millar 2004; Millar and Stewart 2005), parameter…”
NICE DSU Technical Support Document 2: A Generalised Linear Modelling Framework for Pairwise and Network Meta-Analysis of Randomised Controlled Trials (2011, methodology): “For further discussion of priors for variance parameters see Lambert et al.(70) and Spiegelhalter et al.”
Methods to estimate the between‐study variance and its uncertainty in meta‐analysis† (2015, background): “However, the choice of prior distribution is important when the number of studies is small because it may impact on the estimated between-study variance and consequently estimation of the mean treatment effect (Lambert et al., 2005; Senn, 2007).”
A Note Concerning a Selection “Paradox” of Dawid's (2008, background): “…Bayesian estimation works, in the sense of calibrating a Bayesian estimation procedure against a possible constellation of parameters in order to check frequentist properties, as for example in Lambert et al.’s examination of prior distributions for random effect variances (Lambert et al. 2005).”
A re-evaluation of random-effects meta-analysis (2009, background): “The choice of prior distribution, even among those that are intended to be non-informative, can have substantial effects on the results of the analysis (Lambert et al., 2005).”
Conducting indirect-treatment-comparison and network-meta-analysis studies: report of the ISPOR Task Force on Indirect Treatment Comparisons Good Research Practices: part 2. (2011, background): “[22] Lambert PC, Sutton AJ, Burton PR, et al.”
Random effects meta‐analysis of event outcome in the framework of the generalized linear mixed model with applications in sparse data (2010, background): “First, the choice of prior distribution, even among those that are intended to be non-informative, can have substantial effects on the results of the analysis, especially in rare event cases [37, 38].”
Influence of Reported Study Design Characteristics on Intervention Effect Estimates From Randomized, Controlled Trials (2012, background): “9) 809 (41) Mixture of objective and subjective 14 (6.”