Teaching code

4S008983

Teacher

Diego Lubian

Coordinator

Diego Lubian

Credits

9

Language

English

Scientific Disciplinary Sector (SSD)

SECS-P/05 - ECONOMETRICS

Period

Primo semestre (lauree magistrali) dal Oct 2, 2023 al Dec 22, 2023.

Courses Single

Authorized

Lessons timetable Moodle Seminars 0

Learning objectives

Econometrics is the application of statistics and mathematics to economic and business data. In today’s world data are largely available and econometric techniques are crucially important to conducting reliable data analyses in private and public institutions. This course introduces students to regression methods for cross-sectional, time series, and panel data in order to analyze data in business and economics. After presenting the basic theoretical features of each method, the module will offer students the opportunity to practically implement estimation and hypothesis testing techniques in various empirical contexts. The main goal of the course is to provide students with a solid background in econometrics, to stimulate their abilities to apply such techniques on diverse datasets, and to enable them to critically analyze empirical studies in economics, business, and finance.

Prerequisites and basic notions

We require basic knowledge of calculus. The course material relies on prior knowledge of basic statistics and probability theory.

Program

The content of this module will be a balanced mixture of theoretical topics and their empirical application to real dataset. The practical applications involving the use of a statistics/econometrics software will be based on R.
The main topics will be as follow.
Simple linear regression model: model assumptions, ordinary least squares (OLS) and its statistical properties.
Multiple regression model: model assumptions, interpretation of estimates, multicollinearity.
Hypothesis testing: t-tests, F-tests and their distribution.
Dummy variables: definition of intercept and slope dummy variables, and interpretation of coefficients.
Heteroskedasticity: consequence of heteroscedasticity, robust standard errors, testing for heteroskedasticity.
Model misspecification: omitted variable bias, inclusion of irrelevant variables, non-linearity in the regressors.
Endogeneity and instrumental variables: random regressors, definition of endogeneity, instrumental variable estimator (IV), two stages least squares, measurement error, omitted variables and their IV solution, testing for endogeneity.
Simultaneous equations: identification problem, IV solution to simultaneity.
Binary dependent variables: linear probability model, interpretation of coefficients and limitation of this approach. Probit and Logit models.
Time series models: model assumptions, static models, distributed lag models.
Serial correlation: definition of serial correlation, consequences, robust standard errors, testing for serial correlation. Discussion on non stationary time series: consequences of non-stationarity and testing for unit roots.
Panel data analysis: definition of panel dataset, pooled OLS, analysis via differencing, fixed effect and random effects.

Bibliography

Didactic methods

Frontal lectures. During the module we will extensively use the software R to illustrate examples and carry out simple empirical analyses.

Learning assessment procedures

The exam will be a 2-hour written examination. Students will have to answer questions on theory and practice related to the whole program. In addition, students will be asked to carry out a small empirical application by working in groups. Details of the group project will be extensively discussed at the beginning of the course.

Evaluation criteria

To obtain full marks, students should show knowledge of the various econometric methodologies to understand and solve the diverse issues posed by regression models.

Criteria for the composition of the final grade

The final grade will be a weighted average of the marks obtained in the written examination (75%) and in the group project (25%).

Exam language

Inglese