Lecture1presentationsmall.doc

Lecture1 presentation small ECONOMETRIC MODELLING: THE GENERAL TO SPECIFIC PROCEDURE. Some “facts of life”: Economic theory does not tell us much about form content of empirical model Dynamics matter: adjustments are not instantaneous Want to include “relevant” and exclude “irrelevant” variables: but how do we choose? Key themes Principle of parsimony Any model we wish to use must be statistically well-specified. (C) ECONOMIC THEORY Is about equilibrium relationships (1) (D) OUR OBJECTIVES To establish whether the hypothesised economic relationship is supported by empirical evidence. If such support is found: To obtain good estimates of the unknown parameters (1 and (2 To be able to test hypotheses about the unknown parameters To use our estimated regression model for forecasting policy analysis/simulation modelling. [E] THE REQUIREMENTS OF A GOOD ECONOMETRIC MODEL What are the fundamental requirements of an adequate econometric model? (1) THE MODEL MUST BE DATA ADMISSIBLE. It must be logically possible for the data to have been generated by the model. (2) THE MODEL MUST BE CONSISTENT WITH SOME ECONOMIC THEORY. (3) REGRESSORS SHOULD BE (AT LEAST) WEAKLY EXOGENOUS WITH RESPECT TO THE PARAMETERS OF INTEREST. (4) THE MODEL SHOULD EXHIBIT PARAMETER CONSTANCY. (5) THE MODEL SHOULD BE DATA COHERENT. The residuals should be unpredictable from their past history. (6) THE MODEL SHOULD BE ABLE TO ENCOMPASS A RANGE OF ALTERNATIVE MODELS. (F) STATICS AND DYNAMICS: EQUILIBRIUM AND ADJUSTMENTS TO EQUILIBRIUM The regression model counterpart to (1) (assuming time-series data) (2) This is a static econometric model. We need to take account of the possibility of dynamic adjustment processes. Dynamic models are those which contain lagged or leading values of variables, as well as current-dated ones. There are many forms such models can take. Yt = (1 + (2Xt + (3Xt-1 + ut is a distributed