When we are dealing with multivariate observations, the very first questions that come to our mind are:
Where is the data? How is it scattered?
These questions are discussed, different multivariate location and scatter functionals and estimates are presented, and some selected applications are considered. This is an advanced course in statistics for master’s students and doctoral students. Every student is required to give one of the lectures and write and submit a study journal. Attendance is compulsory. There is no exam. The topics of the course are: Location and Scatter Functionals, M-estimates of Location and Scatter, MCD-estimates, Spatial Sign and Rank Based Estimates, Multivariate Location Tests, Autocovariance Matrices and Applications, PCA using different Location and Scatter Estimates, Multivariate Regression Analysis Based on Spatial Signs and Ranks, Scatter Matrix based ICA, Complex-valued Time Series ICA, ICS and Skewness and Kurtosis.
Language of the course: English
Workload: Lectures 24 h, Preparing your own lecture 36 h, Reading and studying the lecture materials 36 h, Writing the study journal 24 h
Assessment Methods and Criteria: Attendance, Lecturing, Study journal
Study Material: Lecture slides
Prerequisites: At leas one master's level statistics or probability theory course
Lecturer: Pauliina Ilmonen, pauliina.ilmonen(a)aalto.fi
Attendance is compulsory! If you are sick and unable to attend a lecture, please contact the lecturer asap.
How to pass this course?
You are expected to
Attend the lectures and be active - this is compulsory - max 12 points.
Give one of the lectures - this is compulsory - max 12 points.
Write and submit a study journal - this is compulsory - max 12 points..
Max total points = 12 + 12 + 12 = 36: You need at least 18 points in order to pass the course.
Grading is based on the total points as follows: at least 18 p -> 1, at least 20 p -> 2, at least 22 p -> 3, at least 26 p -> 4, at least 28 p -> 5.