Course type: elective
Lecturer: Nataša Kejžar
 

For the timeline see Curriculum.

Prerequisites:

• Valid inscription to start the course, positively evaluated home work for taking the exam.

Content (Syllabus outline):
General concepts:

• Formulation of models, estimation of parameters, interpretation
• Interaction, relaxing the linearity assumption
• Explained variation
• Overfitting
• Resampling, validation of models
• Using R in statistical modelling

Logistic regression:

• Fitting the model: maximum likelihood, point and interval estimation of the odds ratio, test statistic, residuals, goodness of fit, influential points
• Interpretation of the model
• Evaluating predictive value of the model
• ROC curves

Objectives and competences:
The aim is for students to learn about strategies of statistical modelling, and to evaluate and validate a model, using logistic regression as an example.

Intended learning outcomes:
Students will be able to fit and interpret models, which will adequately fit the data.