Course content

Statistical learning, multiple linear regression, classification, resampling methods, modell selection/regularization, non-linearity, tree-based methods, neural networks.

Learning outcome

1. Knowledge. The student has knowledge about the most popular statistical models and methods that are used for prediction in science and technology, with emphasis on regression- og classification-type statistical models. 2. Skills. The student can, based on an existing data set, choose a suitable statistical model, apply sound statistical methods, and perform the analyses using statistical software. The student can present, interpret and communicate the results from the statistical analyses, and knows which conclusions can be drawn from the analyses, and what are the caveats.

Learning methods and activities

Lectures, exercises and compulsory works (projects). The assessment is a final written examination (100%), whereas two projects need to be completed as compulsory activities, where at least 60% of the points must be reached to be admitted to the exam. The lectures may be given in English.

Compulsory assignments

• Project

Recommended previous knowledge

The course is based on TMA4240/4245 Statistics, or equivalent. Good understanding of linear algebra (matrix methods) and optimization.

Course materials

James, G., Witten, D., Hastie, T., Tibshirani, R. "An Introduction to Statistical Learning with Applications in R", Springer. Additional literature will be announced at the start of the course. 2nd Edition.

Credit reductions