The Robustness of Estimators for Dynamic Panel Data Models to Misspecification

The transition from economic theory to a testable form invariably involves the use of certain "simplifying assumptions". However, if these are not valid, misspecified model result. This paper considers consistent estimation of the dynamic panel model which often forms the basis of testable economic hypotheses. These estimators are similarly based on certain "simplifying assumptions", which appear to be often untenable in practice. In light of this, unlike previous studies, the performance of these estimators is analysed in scenarios where these assumptions have been deliberately violated. The wide range of potential parameter estimates and conflicting inference amongst them, is illustrated by a investigation into a simple and most likely misspecified, relationship between savings and output growth