Work in R - specific tasks

Programming and visualization tools in R for the needs of econometric applications. Basic work with data, data structures, import, export and data transformation. Specific data problems (missing, remote and contaminated observations).

Linear regression

Classical linear regression model. Model formulation, cross-sectional and dynamic models, special regressors (eg dummy or trend variables). Estimators and their properties - properties on final selections and asymptotic properties. OLS, GLS, MLE, robust estimators. Testing and consequences of heteroskedasticity and autocorrelation. Measurement of multicollinearity. Applications in macroeconomic and microeconomic fields.

Discrete and limited explanatory variables

Models with special explanatory variables (eg linear probability model, censored and truncated regression models, logit, probit and tobit model). ML estimators. Specification issues. Applications in scoring models.

Multi-equation models and panel data

General formulation of the system. SUR system. Models for panel data. System of simultaneous equations. Identification and estimators.

Time series

One-dimensional time series: Decomposition methods (seasonality, trends). Box-Jenkins methodology. Stationary. Applications in portfolio theory. Multidimensional time series: Generalization of one-dimensional methods. Vector autoregression. Granger causality. Cointegration and error correction models.