Title :
Bayesian analysis of panel data based on MCMC
Author_Institution :
Sch. of Manage., Tianjin Univ., Tianjin
Abstract :
This paper developed a Bayesian method to analyze panel data by using the MTAR model. The key feature of the proposed method is that it can perform posterior estimation and interesting multiple hypotheses testing simultaneously by using a simple MCMC algorithm. It takes the advantages of Bayesian multiple hypotheses testing while avoiding the computational difficulties in posterior inference. The illustrative example of test of optimal debt/total assets ratio shows practical usefulness of the propose method.
Keywords :
Bayes methods; autoregressive processes; parameter estimation; Bayesian analysis; Bayesian method; Bayesian multiple hypotheses testing; panel data; posterior parameter estimation; threshold autoregressive model; Bayesian methods; Conference management; Cybernetics; Data analysis; Fluctuations; Inference algorithms; Machine learning; Performance evaluation; Testing; Time series analysis; MCMC; Posterior Estimation; Threshold Autoregressive;
Conference_Titel :
Machine Learning and Cybernetics, 2008 International Conference on
Conference_Location :
Kunming
Print_ISBN :
978-1-4244-2095-7
Electronic_ISBN :
978-1-4244-2096-4
DOI :
10.1109/ICMLC.2008.4620580
