Modèles de régression et Statistical Softwares

Part 1. Linear regression (least squares estimation, matrix notation, variance estimation, exact and asymptotic inference on the regression parameter, weighted and generalized least squares estimations)
Part 2. Introduction to the use of a statistical software: tools for exploratory analysis, variable manipulation, plots. Applied regression analysis (analysis and goodness-of-fit procedures)

After following this teaching unit, a student will be able to:
- explain why regression models are of interest
- perform a regression analysis that is suitable to the context
- conduct, on the computer, a regression analysis and related analyses

Theoretical lectures
Project/howeworks
Use of statistical softwares on various examples

Contribution to the teaching profile

• Being able to analyze rigorously and with critical thinking a data set- Chosing the suitable actuarial models and techniques for a given problem- Analyzing with rigor and critical thinking the results obtained- Becoming independent enough to identify, collect and analyze the data needed to solve a given problem- Using a clear and rigorous language- Writing in a rigorous and concise way reports presenting a problem, the models and techniques used to address it, and the results obtained- Present verbally in a clear and concise way the results of an analysis

References, bibliography, and recommended reading

Ravishanker, N., and D. K. Dey (2001). A first course in linear model theory, Chapman & Hall.
Dobson, A. J. (2001). An introduction to generalized linear models, Chapman & Hall.

Course notes

• Syllabus
• Université virtuelle

Contacts

Davy Paindaveine <Davy.Paindaveine@ulb.be>
Toufik Zahaf <tzahaf@ulb.ac.be>

Campus

Plaine

Method(s) of evaluation

• written examination
• Project

written examination

Project

Part 1: exam of a theoretical nature (the exam will offer both open and closed questions)
Part 2: project/Writing of a report
 

Mark calculation method (including weighting of intermediary marks)

The exam in part 1 and the project in part 2 each contribute to 50% of the final grade.

For the 50% associated with part 2, the various components of the project will be evaluated as follows:
- Exploratory analysis: /4 points
- Regression analysis: /8 points
- Quality of the report: /4 points
- Knowledge in statistics (on the basis of the oral presentation): /4 points

Grades from Part 1 or Part 2 that are larger than or equal to 10/20 are pushed to the next session and/or to the next academic year.

Language(s) of evaluation

• french
• (if applicable english )