In this second course in Statistics the basic elements of Probability and Statistical Inference are extended and broadened. The main part of the course concerns the models for studying multivariate dependencies based on linear regression for continuous responses and on logistic models for binary responses. An essential part of the course is the study of several data sets providing real illustrations for the discussed methods.
Giovanni Marchetti (2012). Introduzione ai Modelli Statistici. Firenze, Dispense del Dipartimento di Statistica, Informatica, Applicazioni. See http://local.disia.unifi.it/gmm/.
Learning Objectives
We discuss the basic models for the study of multivariate dependencies. A central theme is the multiple regression model.
Special care is devoted to the building of models in the presence of mixed continuos an binary variables, including possibly quadratic terms and interactions.
The logistic regression model for binary responses is also treated. The course gives also an introduction to conditional independence and multivariate models, with several illustrations and applications from science, economics and technology.
Prerequisites
Analisi I e II, Geometria I, Calcolo delle Probabilita’ e Statistica
Teaching Methods
Standard lectures and computer practicals on real data using the R language
Further information
Frequency of lectures, practice and lab: Recommended
Office hours:
during the course: Wednesday: 14:00 to 16:00
or by appointment
Contact:
Dipartimento di Statistica Informatica e Applicazioni (DiSIA) in viale Morgagni, 65, 50134 Firenze (Ex-Farmacologia stanza 1/25)
email: giovanni.marchetti@disia.unifi.it
Type of Assessment
Homework is assigned each week. The final exam is oral.
Course program
1. Random sampling
2. The Gaussian model
3. Basic inference
4. Models for groups comparison
5. Correlation and independence
6. Simple linear regression
7. Likelihood-based inference
8. Simple linear logistic model
9. Multiple linear regression
10. Model building
11. General linear logistic models