Problems in applying dynamic panel data models: Theoretical and empirical findings

The objective of this paper is twofold: First, the applicability of a widely used dynamic model, the autoregressive distributed lag model (ARDL), is scrutinized in a panel data setting. Second, Chile's development of market shares in the EU market in the period of 1988 to 2002 is then analyzed in this dynamic framework, testing for the impact of price competitiveness on market shares and searching for estimation methods that allow dealing with the problem of inter-temporal and cross-section correlation of the disturbances. To estimate the coefficients of the ARDL model, FGLS is utilized within the Three Stage Feasible Generalized Least Squares (3SFGLS) and the system Generalized Method of Moments (system GMM) methods. A computation of errors is added to highlight the susceptibility of the model to problems related to underlying model assumptions.