Teacher(s)
Language
English
> French-friendly
> French-friendly
Prerequisites
English at the medium level and concepts and tools equivalent to those taught in the course LSTAT2014
Main themes
- Introduction to the general linear model - Multiple univariate regression (selection of variables, model validation, multicollinearity, outlier detection, inference concerning regression coefficients, error variance,...) - Univariate analysis of variance (one or more factors, balanced or non-balanced design, fixed, mixed or random effects model, inference concerning main effects, interactions, error variance,...) - Multivariate regression and multivariate analysis of variance
Learning outcomes
At the end of this learning unit, the student is able to : | |
1 |
By the end of this course the student will be familiar with the main linear models that are often encountered in statistics, and, by making use of computer packages, the student will be able to solve real data problems. The course stresses more the methodology, the interpretation, and the mechanisms behind linear models, and less the theoretical and mathematical aspects. |
Content
1 The model, specification and interpretation
2 Estimation and geometry
3 Statistical properties of OLS
4 Maximum likelihood estimator
5 Inference and hypothesis tests
6 Multicollinearity
7 Discrete variables and Analysis of Variance
8 Variable selection
9 Heteroskedasticity, autocorrelation
10 Diagnostics (outliers, influential observations)
11 Panel data
2 Estimation and geometry
3 Statistical properties of OLS
4 Maximum likelihood estimator
5 Inference and hypothesis tests
6 Multicollinearity
7 Discrete variables and Analysis of Variance
8 Variable selection
9 Heteroskedasticity, autocorrelation
10 Diagnostics (outliers, influential observations)
11 Panel data
Teaching methods
The course consists of main lectures, which provide the main theoretical material but also give many practical examples, and exercise sessions that teach how to implement the methods in a common statistical programming package. The exercise sessions will also help the students to work on their individual projects.
Evaluation methods
A written exam (60%) and an individual project on a real data set (40%).
Bibliography
Kutner, Nachtsheim, Neter and Wasserman, Applied Linear
Statistical Models, McGraw-Hill, Fifth edition (2005)
Davidson and McKinnon, Econometric Theory and
Methods, Oxford University Press (2004)
other references will be given in class.
Statistical Models, McGraw-Hill, Fifth edition (2005)
Davidson and McKinnon, Econometric Theory and
Methods, Oxford University Press (2004)
other references will be given in class.
Teaching materials
- matériel sur moodle
Faculty or entity
Programmes / formations proposant cette unité d'enseignement (UE)
Title of the programme
Sigle
Credits
Prerequisites
Learning outcomes
Master [120] in Data Science : Statistic
Master [120] in Biomedical Engineering
Master [120] in Statistics: Biostatistics
Master [120] in Mathematics
Master [120] in Statistics: General
Master [120] in Chemistry and Bioindustries
Approfondissement en statistique et sciences des données
Master [120] in Mathematical Engineering
Master [120] in Data Science Engineering
Minor in Statistics, Actuarial Sciences and Data Sciences
Certificat d'université : Statistique et science des données (15/30 crédits)
Master [120] in Data Science: Information Technology