All programmes > Linear Models

Linear Models (B-KUL-G0Z50A)

6 ECTSEnglish39 First termCannot be taken as part of an examination contract
Alonso Abad Ariel (coordinator) |  Alonso Abad Ariel |  Neyens Thomas
POC Master in statistiek

Aims

This course aims to provide thorough insight into linear modeling. Students also gain practical experience of analysing data sets with modern statistical software.   

Previous knowledge

Students should have good knowledge about calculus (differentiation, logarithmic and exponential functions), linear algebra (simple matrix operations, inverses, eigenvalues and eigenvectors) and the basic concepts of probability and statistics (random variables, confidence intervals, hypothesis testing, p-values).

Preliminary requirements: Students should have had an introductory course in statistics, a calculus course and a course containing at least an introduction to linear algebra. The first condition can be met with, for example, the course “Fundamental concepts of statistics”.

Identical courses

This course is identical to the following courses:
G0Z50B : Linear Models

Is included in these courses of study

Activities

6 ects. Linear Models (B-KUL-G0Z50a)

6 ECTSEnglishFormat: Lecture39 First term
Alonso Abad Ariel |  Neyens Thomas
POC Master in statistiek

Content

  • Simple linear regression
  • The general linear model
  • Multiple linear regression: inference
  • Polynomial regression
  • Categorical predictors
  • Transformations
  • Variable selection methods
  • Multicollinearity and its remedials
  • Outliers and influential observations
  • One-way ANOVA: F-test
  • Factor levels effects and multiple comparison
  • Diagnostics and remedial measures
  • Model reformulation and regression approach
  • Two-way ANOVA: Equal sample size
  • Two-way ANOVA: Unequal sample size
  • Estimable Functions
  • Type I, II and III Sum of Squares

Course material

Slides
Course notes
Suggested handbook: Applied Linear Statistical Models, 5th Edition, Kutner et al. (2005)

Is also included in other courses

Evaluation

Evaluation: Linear Models (B-KUL-G2Z50a)

Type : Partial or continuous assessment with (final) exam during the examination period
Description of evaluation : Oral, Paper/Project, Report
Type of questions : Open questions
Learning material : Course material, Calculator, Reference work

Explanation


  • The students will have to do a project divided in tutorial groups. Each tutorial group has to write a report with a detailed discussion of the analysis. A report should contain no more than 3 sheets in total, including the title page, i.e., if you print your report it should not exceed 3 sheets two sided. Every page should only have one column of text.  Please use an A4 page format, a times new roman 12 font and a 1.2 spacing between lines. The title page of the report should contain the number of the group and a list with the names and student numbers of all the members of the group. Please note that these reports are part of your evaluation, thus follow these instructions very carefully. The project will account for 5 points and the exam for 15 points.
  • The students that follow the on-campus program will have a written multiple choice exam that will account for 15/20 points. In the written multiple choice exam correction for guessing will be used. If students cannot attend the regular exam for a justified reason, they may be given an oral or written exam at a time specified by the professor.
  • The project is a compulsory task and an integral part of the exam. Students who do not complete or contribute to the project with their group will receive an NA for both the regular and second-chance exams. If a group does not complete the project task or fails to submit it on time, all members of the group will receive an NA for both the regular and second-chance exams.

 

 

Information about retaking exams

The grade of the project will be carried forward to the second chance exam