 | |
| STAT1216-1 | Linear models and statistical modelling
 |
|
| Duration : | 36h Pr, 36h Th |
|
| Credits/ECTS : |
|
|
| Holder(s) : | Rodolphe Palm |
|
| Language : | French language |
|
| Course contents : | Linear regression
- simple and multiple regression
- variable selection
- examining residuals and checking assumptions
- weighted regression and variable transformation
Simple, multiple and partiel correlation
Comparison of two or several populations
Principal component regression and PLS
Ridge regression
Non linear regression
Logistic regression
Robust and non parametric regression |
|
| Course objective : | To revise and complete the basic elements of simple and multiple linear regression, to learn advanced statistical methods to model a response variable as a function of one or several predictors.
After completing the course the student is expected to
- model correctly the relationship between a dependent variable and one or several explanatory variables,
- check fitted models,
- compare models related to two or several populations. |
|
| Prerequisites : | Basic skills in applied statistics, for example :
-- STAT1207-1- Applied statistics
Basic skills in computer science, for example :
-- HULG0149-1- Office automation
or -- INFO2037-1- Introduction to computer science |
|
| Organization : | Lectures : 36h
Practical Works : 36h |
|
| Written notes : | - Syllabus.
- Reference book : DRAPER N.R., SMITH H. [1998]. Applied regression analysis. New York, Wiley, 706 p. |
|
| Assessment : | Written examination (100%) |
|
| Contacts : | Palm, Rodolphe (Professeur)
Enseignement et recherche
081 62 24 79
Rodolphe.Palm@ulg.ac.be |
|
