This course presents the multiple linear econometric model as well systems of linear equations, estimated simultaneously. The course gives a thorough presentation of the above models based on linear algebra techniques. In particular, the first part of the course presents the multiple linear econometric model under its classical assumptions and discuss issues of inference and estimation of its coefficients based on the least squares (LS) estimator. It also shows how to use this model to produce out-of-sample predictions. In its second part, the course presents statistical properties of LS estimator for large samples under less restrictive assumptions than the classical ones. In this section, it presents diagnostic tests for the econometric problems of autocorrelation and heteroscedasticity which are often appeared in practice. It also presents the generalized least squares (GLS) estimator for the estimation of linear econometric models which suffer from the above two problems. This part of the course also presents the estimation method of maximum likelihood.In the third part, the course presents the method of moments estimator and the instrumental variables estimator which corrects for the problem of endogeneity of regressors. Next, it presents linear systems of equations and discuss problems of identification of their structural parameters and estimation, with the method of two stages least squares. For students practice, the course provides tutorials for exercise solving and applications of linear econometric models to the Greek economy based on an econometric package.
Indicative Course Prerequisites: Applied Econometrics, Mathematics for Economists ΙΙ, Statistics ΙΙ