05555oam 22009734 450 991081232000332120200520144314.01-4623-7192-21-4527-1274-397866128429551-4518-7221-61-282-84295-1(CKB)3170000000055239(EBL)1608239(SSID)ssj0000941859(PQKBManifestationID)11614171(PQKBTitleCode)TC0000941859(PQKBWorkID)10971226(PQKB)10055765(OCoLC)608248516(IMF)WPIEE2009074(MiAaPQ)EBC1608239(IMF)WPIEA2009074(EXLCZ)99317000000005523920020129d2009 uf 0engur|n|---|||||txtccrLimited Information Bayesian Model Averaging for Dynamic Panels with Short Time Periods /Alin Mirestean, Charalambos Tsangarides, Huigang Chen1st ed.Washington, D.C. :International Monetary Fund,2009.1 online resource (45 p.)IMF Working PapersDescription based upon print version of record.1-4519-1656-6 Includes bibliographical references.Contents; I. Introduction; II. Model Uncertainty in the Bayesian Context; A. Model Selection and Hypothesis Testing; B. Bayesian Model Averaging; C. Choice of Priors; III. Limited Information Bayesian Model Averaging; A. A Dynamic Panel Data Model with Endogenous Regressors; B. Estimation and Moment Conditions; C. The Limited Information Criterion; IV. Monte Carlo Simualtions and Results; A. The Data Generating Process; B. Simulation Results; V. Conclusion; References; Tables; 1. Posterior Probability of the True Model; 2. Posterior Probability Ratio of True Model/Best among the Other Models3. Probability of Retrieving the True Model4. Model Recovery: Medians and Variances of Posterior Inclusi; 5. Model Recovery: Medians and Variances of Estimated Paramet; 6. Posterior Probability of the True Model (Non-Gaussian Case); 7. Posterior Probability Ratio: True Model/best among the Other Models (Non-Gaussian Case); 8. Probability of Retrieving the True Model (Non-Gaussian Case); 9. Model Recovery: Medians and Variances of Posterior Inclusion Probability for Each Variable (Non-Gaussian Case); 10. Model Recovery: Medians and Variances of Estimated Parameter Values (Non- Gaussian Case)Appendix A Figures1. Posterior Densities for the Probabilities in Table 1; 2. Posterior Densities for the Probabilities in Table 2; 3. Box Plots for Parameters in Table 5; 4. Posterior Densities for the Probabilities in Table 6; 5. Posterior Densities for the Probabilities in Table 7; 6. Box Plots for Parameters in Table 10Bayesian Model Averaging (BMA) provides a coherent mechanism to address the problem of model uncertainty. In this paper we extend the BMA framework to panel data models where the lagged dependent variable as well as endogenous variables appear as regressors. We propose a Limited Information Bayesian Model Averaging (LIBMA) methodology and then test it using simulated data. Simulation results suggest that asymptotically our methodology performs well both in Bayesian model selection and averaging. In particular, LIBMA recovers the data generating process very well, with high posterior inclusion probabilities for all the relevant regressors, and parameter estimates very close to the true values. These findings suggest that our methodology is well suited for inference in dynamic panel data models with short time periods in the presence of endogenous regressors under model uncertainty.IMF Working Papers; Working Paper ;No. 2009/074Panel analysisBayesian statistical decision theoryBayesian Analysis: GeneralimfBayesian inferenceimfBayesian modelsimfComputer Programs: GeneralimfData capture & analysisimfData Collection and Data Estimation MethodologyimfData ProcessingimfData processingimfEconometric modelsimfEconometrics & economic statisticsimfEconometricsimfElectronic data processingimfEstimation techniquesimfEstimationimfPanel analysis.Bayesian statistical decision theory.Bayesian Analysis: GeneralBayesian inferenceBayesian modelsComputer Programs: GeneralData capture & analysisData Collection and Data Estimation MethodologyData ProcessingData processingEconometric modelsEconometrics & economic statisticsEconometricsElectronic data processingEstimation techniquesEstimation332.152Mirestean Alin1659946Chen Huigang1599330Tsangarides Charalambos1462110DcWaIMFBOOK9910812320003321Limited Information Bayesian Model Averaging for Dynamic Panels with Short Time Periods4014884UNINA