A Bayesian Approach to Model Uncertainty, Issues 2004-2068
International Monetary Fund, Apr 1, 2004 - Business & Economics - 21 pages
This paper develops the theoretical background for the Limited Information Bayesian Model Averaging (LIBMA). The proposed approach accounts for model uncertainty by averaging over all possible combinations of predictors when making inferences about the variables of interest, and it simultaneously addresses the biases associated with endogenous and omitted variables by incorporating a panel data systems Generalized Method of Moments estimator. Practical applications of the developed methodology are discussed, including testing for the robustness of explanatory variables in the analyses of the determinants of economic growth and poverty.
What people are saying - Write a review
We haven't found any reviews in the usual places.
Model Uncertainty and BMA
Statistics for the Robustness Analysis
Other editions - View all
account for model approximation Assumption A3 asymptotic Bayes factor Bayesian counterpart Bayesian framework Bayesian Information Criterion Bayesian Method Bayesian Model Averaging Brock and Durlauf calculate classical GMM context of GMM Covariance Matrix Csiszar defined Denote density Econometrica Econometrics endogeneity equal inclusion equations error explicitly account GMM estimator growth h(xt Hansen Heteroskedasticity Hoeting hypothesis testing inference Kass and Raftery Lemma LIBMA approach likelihood function Limited Information Bayesian limited information likelihood limited information procedure logn loss function marginal likelihood Method of Moments Miller minimization model Mk model selection Moments Estimation number of regressors possible regressors post-data posterior distribution posterior inclusion probability posterior mean posterior model probabilities posterior odds ratio posterior probabilities prior odds ratio problem of model qT(xT\Mk Radon-Nikodym derivative right hand side robustness analysis Sala-i-Martin Schwarz criterion Section set of possible solution T/oc true probability measure variable vector weighing matrix weights Zellner zero