32443 - Statistical Methods for Data Analysis

Learning outcomes

The aim of the course is to make the students able to coherently use some statistical techniques for the analysis of multivariate data. The procedures covered in the course include:

- Introduction to multivariate analysis
- Multiple regression;
- Chi-square test
- T-test
- Principal components analysis
- Factor analysis.

Course contents

1. Introduction to multivariate statistical analysis.
2. Multiple linear regression: model specification, parameters estimation, residual analysis.
3. Chi square test for independence
4. T-test to compare means
5. Principal components analysis.
6. Factor analysis.

Readings/Bibliography

Inference:
Borra, Di Ciaccio (2014) Metodologie per le scienze economiche e sociali, 3^ edizione, Milano, McGraw-Hill education: chapters 8-14

Regression model:
Borra Di Ciaccio (2014): chapters 16-17

Multiple regression model (on-line):
Available on EOL platform

Chi-square test and T-test:
Slides online (EOL platform)

Principal components analysis and Factor analysis:
Slides online (EOL platform)

Video of the lectures are available on EOL platform

Teaching methods

- Lectures, during which theoretical concepts as well as statistical techniques are explained and illustrated by examples.
- Tutorial sessions in computer laboratory.

Assessment methods

Depending on the sanitary provisions, the methods of examination may vary.
"Traditional" assessment:
Written test (mandatory) and oral examination (optional).

"Covid" assessment:
You can choose between two alternatives:
- Teamwork (max 3 people): the team is given a series of problems to solve. Within the estabilished time, the team will have to return a document containing the required analyzes and comments.
The team can also be composed by one candidate and will obviously be proportionate.
- Online test: each candidate carries out a multiple choice test on the EOL platform. 20 questions in 45 minutes.

In both cases an oral test will follow.

Point system:
<18 not sufficient
18-23 sufficient
24-27 good
28-30 very good
30 e lode excellent

Office hours

See the website of Alessandro Lubisco