Unit STATISTICAL LEARNING FOR DATA SCIENCE

Course
Finance and quantitative methods for economics
Study-unit Code
A003083
Location
PERUGIA
Curriculum
Data science for finance and insurance
Teacher
David Aristei
Teachers
  • David Aristei
Hours
  • 42 ore - David Aristei
CFU
6
Course Regulation
Coorte 2023
Offered
2024/25
Learning activities
Caratterizzante
Area
Matematico, statistico, informatico
Academic discipline
SECS-S/01
Type of study-unit
Opzionale (Optional)
Type of learning activities
Attività formativa monodisciplinare
Language of instruction
English
Contents
The course is organized in four main parts:
1) Models with binary dependent variables
2) Multi-response models
3) Limited dependent variable models
4) Linear regression models for longitudinal data
Reference texts
Main textbook:
Verbeek, M., A guide to Modern Econometrics, Fourth Edition, Wiley, 2012.

Additional suggested textbooks:
- Greene, W., Econometric Analysis, 7th Ed., Prentice Hall, 2012.
- Wooldridge, J.M., Econometric Analysis of Cross Section and Panel Data, 2nd Ed. Mit Press, 2010.
- Cameron, A.C. e Trivedi, P.K., Microeconometrics: Methods and Applications, Cambridge Press, 2005.

Class slides and materials used during the laboratory sessions will be distributed through the UniStudium e-larning platoform
Educational objectives
This course will introduce students to the most relevant methods of advanced econometric analysis and to their use in estimating economic relationships with cross-sectional and panel data.
Specific attention will be also devoted to on empirical applications. Every week there will be a computer laboratory session in which students will apply concepts learned during the theoretical lectures, using the econometric software STATA.
Prerequisites
This course requires knowledge of inferential statistics and multiple regression analysis, which should be gained during the first year of the second cycle degree course.
Teaching methods
Face-to-face lessons completed with practical laboratory activities
Other information

Learning verification modality
Written examination
Extended program
1) Models with binary dependent variables
The linear probability model.
Logit and probit models.
Specification issues: binary choice model with heteroscedasticity
Binary choice models with endogenous regressors

2) Multi-response models
Ordered Response Models: ordered logit and ordered probit models.
Multinomial models: the multinomial logit model.

3) Limited dependent variable models
Censored and truncated dependent variables.
Truncanted regression model
The Standard Tobit model
Extensions of the standard Tobit model: the Tobit II model
Sample selection bias

4) Linear regression models for longitudinal data
The static linear model.
The fixed- effect model.
The random-effects model.
Specification tests.
Obiettivi Agenda 2030 per lo sviluppo sostenibile
4, 8, 9, 12
Condividi su