A Comparative Study About Issues and Challenges In Evaluating Econometric Models |

ABSTRACT:

Econometrics, in its longhistory, has been and continues to be an important branch not only in generaleconomics (macro and micro), but also in specialized fields in the area ofeconomics, such as financial and spatial economics. This paper surveys somerecent developments related to the specification and estimation of econometricmodels widely used in applied research. Even though we lay emphasis on timeseries models and their application in financial and spatial econometrics,additional topics, such as limited dependent variable models and simultaneousequation systems, are also reviewed in the paper. However, it should be emphasizedthat the survey is not unified in the sense that it does not provide anexhaustive review of the development of econometrics through its long history. Standard econometric model selection methods arebased on four conceptual errors: parametric vision, the assumption of a truedata generating process, evaluation based on fit, and ignoring the impact ofmodel uncertainty on inference+ Instead, econometric model selection methodsshould be based on a semiparametric vision, models should be viewed as approximations,models should be evaluated based on their purpose, and model uncertainty shouldbe incorporated into inference methods.