Applied Multivariate Statistics

Spring semester 2018

General information

Lecturer Fabio Sigrist
Assistant Niklas Pfister, Emilija Perkovic
Lectures Mon 8-10 HG F 3 >>
Exercises Mon 15-17 (every second week) HG E 1.2 >>
Course catalogue data >>

Course content

Multivariate statistics analyzes data on several random variables simultaneously. This course introduces the basic concepts and provides an overview of classical and modern methods of multivariate statistics including visualization, dimension reduction, supervised and unsupervised learning for multivariate data. An emphasis is on applications and solving problems with the statistical software R.

Literature

  • "An Introduction to Applied Multivariate Analysis with R" (2011) by Everitt and Hothorn
  • "An Introduction to Statistical Learning: With Applications in R" (2013) by Gareth, Witten, Hastie and Tibshirani
  • "Introductory Statistics with R" (2008) by Dalgaard

Additional information

Examples in the lecture and hints as well as solutions to the exercises will be based on the statistical software R. This is a freely available open source program that works on all platforms and has become worldwide standard for data analysis. It can be downloaded from CRAN. A R Tutorial can be found here.

Exam

There will be a 120-minute written exam during the regular ETH exam sessions. The exam is a "closed book" exam, a simple pocket calculator with no communication capability is permitted. The exam covers all topics which were discussed and/or applied during either the lectures or the exercises. Upon passing the exam, the course will be awarded 5 ECTS credit points.

PhD students who would like to obtain credit points but do not need to take the exam and obtain a grade need to sign up with the lecturer at the beginning of the semester and hand in 5 well-solved exercises.

Announcements

  • Februar 3, 2017:
    Beginning of lecture: Monday, 19/02/2018 (Exercises start on 26/02/2018 at 15:15).

Course materials

The schedule is subject to minor modifications. The slides will be uploaded before each lecture.

Week Topic
Week 1 Introduction & visualization
Week 2 Visualization & outliers
Week 3 Principal component analysis 1
Week 4 Principal component analysis 2
Week 5 Multidimensional scaling
Week 6 Factor analysis
Week 7 Cluster analysis
Week 8 Classification 1: discriminant analysis & logistic regression
Week 9 Extending univariate methods
Week 10 Classification 2: trees & random forest
Week 11 Models for repeated measures data
Week 12 Selection of additional machine learning methods

Literature:

  • Everitt and Hothorn (2011), An Introduction to Applied Multivariate Analysis with R, Springer
  • James, Witten, Hastie, and Tibshirani (2013), An Introduction to Statistical Learning, Springer
  • P. Dalgaard (2008), Introductory Statistics with R, Springer

Exercise classes

Exercise classes are held approx. every second week starting from 26/02/2018 in HG E 1.2. Please install R and RStudio and bring your laptop to the exercise classes, if possible.

Series and solutions

There is no testat requirement for students who take the exam. PhD students who do not take the exam but would like to obtain a testat should send their solutions to the assistants by email no later than one week after the discussion in the exercise class.

Exercises Solutions Date of exercise class
Series 1 Solutions 1 26.02.2018
Series 2 Solutions 2 12.03.2018
Series 3 Solutions 3 26.03.2018
Series 4 Solutions 4 23.04.2018
Series 5 Solutions 5 07.05.2018
Series 6 Solutions 6 28.05.2018