Statistics - KAEL7M05

Goal(s)

At the end of the course, students should be able to:

• describe data using statistical summaries and also propose appropriate visualisations.
• reduce multidimensional data and analyse the principal components (principal component analysis) or discriminant components (linear discriminant analysis)
• estimate the mean (and the proportion) of a random variable in the population from a sample of measurements, and give the estimate by confidence interval.
• carry out parametric hypothesis tests: comparison of a mean with a norm, of two means, of several means, etc.

Content(s)

The course is divided into 2 main parts, the first dealing with descriptive statistics and the second with inferential statistics.
In the first part, we will review the classic descriptive statistics used to describe 1 and 2 variables (qualitative and quantitative). We will also look at multidimensional descriptive statistics, with principal component analysis and linear discriminant analysis in its descriptive (and not predictive) approach.
The second part of the course covers confidence interval estimation of the mean (proposition) as well as parametric hypothesis tests comparing a mean (proportion) to a standard, 2 means (proportions) and several means using analysis of variance.
The tutorials are given on a PC using Python and the Numpy and Pandas libraries (seaborn will also be covered), as well as the scipy.stats library.

Year 3 Maths course, particularly the chapter on probability and random variables. Notions of matrix calculations and diagonalization are also required.