In the context of an autoregressive panel data model with ﬁxed effect, we examine the relationship between consistent parameter estimation and consistent model selection. Consistency in parameter estimation is achieved by using the tansformation of the ﬁxed eﬀect proposed by Lancaster (2002). We ﬁnd that such transformation does not necessarily lead to consistent estimation of the autoregressive coeﬃcient when the wrong set of exogenous regressors are included. To estimate our model consistently and to measure its goodness of ﬁt, we argue for comparing diﬀerent model speciﬁcations using the Bayes factor rather than the Bayesian information criterion based on the biased maximum likelihood estimates. When the model uncertainty is substantial, we recommend the use of Bayesian Model Averaging. Finally, we apply our method to study the relationship between ﬁnancial development and economic growth. Our ﬁndings reveal that stock market development is positively related to economic growth, while the eﬀect of bank development is not as signiﬁcant as the classical literature suggests.
Online Research @ Cardiff (http://orca.cf.ac.uk/42902/1/E2009_5.pdf)