Classical multiple linear regression model: ordinary least squares (OLS), goodness of fit and analysis of variance. Finite sample properties of the OLS estimator: unbiased estimation, variance of the OLS estimator and the Gauss Markov theorem. Estimation of the variance of the least square estimator. Normality assumptions and basic statistical inference. Data problems: multicollinearity and missing observations. Large sample properties of the OLS estimator: consistency, asymptotic normality, asymptotic efficiency. Instrumental Variables and Hausman’s specification test. Inference and Prediction. Tests for structural change: dummy variables, partitioned regression. Specification analysis and model selection: irrelevant variables and omission of relevant variables. Nonspherical disturbances and generalized regression model: GLS and FGLS. Heteroskedasticity: inefficiency of OLS, estimated covariance matrix of the parameters, Generalized Method of Moments (GMM), estimation of the heteroskedastic regression model, testing for heteroskedasticity. Serial Correlation: disturbance processes, testing for autocorrelation; models with lagged variables.