UNIGE document Doctoral Thesis
previous document  unige:104670  next document
add to browser collection

Validity and accuracy of posterior distributions in Bayesian statistics

Turbatu, Laura Florina
Defense Thèse de doctorat : Univ. Genève, 2018 - GSEM 52 - 2018/04/17
Abstract In this thesis I investigate the validity and the accuracy properties of the posterior quantiles in Bayesian statistics when replacing the parametric likelihood with the Cressie-Read empirical likelihoods based on a set of unbiased M-estimating equations. At first order I study the validity of the empirical posterior distribution derived from the pseudo-likelihood constructed with profiled weights and estimated at a minimum distance from the empirical distribution in the Cressie-Read family of divergences, indexed by γ. The bias in coverage of the resulting empirical posterior quantile is inversely proportional to the asymptotic efficiency of the estimator corresponding to the set of M-estimating functions. By comparing different members of the Cressie-Read family of empirical likelihoods for models in the exponential family, I establish a hierarchy in the accuracy of the quantile function of the empirical posterior distribution depending on the index parameter γ.
Keywords Empirical likelihoodExponential tiltingPosterior quantilesValidityAccuracyHigher-order propertiesCressie-Read family of divergencies
URN: urn:nbn:ch:unige-1046703
Full text
Thesis (1.7 MB) - public document Free access
(ISO format)
TURBATU, Laura Florina. Validity and accuracy of posterior distributions in Bayesian statistics. Université de Genève. Thèse, 2018. doi: 10.13097/archive-ouverte/unige:104670 https://archive-ouverte.unige.ch/unige:104670

404 hits



Deposited on : 2018-05-28

Export document
Format :
Citation style :