The objective of the course is to provide students with a thorough coverage of the classical econometric theory and with the computational tools to be
used in the empirical analyses. The program varies with the students background, but generally includes the following topics: the classical
regression model, relaxing the assumptions of the classical model, time series econometrics and simultaneous equation models.
1. Interpolation with Ordinary Least Squares Method (OLS)
2. Simple and K-variables Linear Regression Model Basic assumptions, OLS estimation.
Algebraic Properties of the estimates, Statistical Properties of the estimates, the Gauss-Markov theorem, The Coefficient of determination Unbiased estimation of .
The normality assumption, distributions of quadratic forms. Independence between quadratic forms, independence between a quadratic form and a linear form.
test-t, test-F, alternative forms of the test-F, test of hypothesis (linear restrictions). Regression and forecasting.
3. Further results on the regression model: Restricted Least Squares , structural changes, Dichotomous variables (dummy variables ), multicollinearity.
4. Generalized Least Squares (GLS), Non spherical disturbances and OLS estimates , Generalized Least
Squares (GLS). Equivalence between GLS and OLS on transformed variables.
5. Rudiments of asymptotic theory: Convergence in probability and convergence in distribution. OLS estimation of dynamic models: the instrumental variables
method (IV). Delta-Method
6. Introduction to linear simultaneous equations models: Structural form and reduced form, simultaneous equations models
and inconsistency of OLS estimation. The identification problem. Single equation estimation methods in simultaneous equations models: Indirect Least Squares,
Two Stage Least Squares (TSLS), Instrumental Variables.
7. Nonlinear Least Squares, ML estimation in linear and nonlinear models