Statistical computing environment

The aim of the studying block is to provide an overview of taught methods and clarification of prerequisite knowledge of statistics. Furthermore, acquaintance with computing environments that will be used in the course by the implementation of basic statistical tasks.

Descriptive statistics

The aim of the block is to explain the methods of descriptive statistics suitable for the description of data and their testing in a computing environment.

Problems in data

The aim of this block is to describe the methods of diagnosis and solution of the most common problems in the data, which are associated with dimensionality, outliers, inconsistencies, missing values, non-numerical data, etc.

Data transformation

Transformations according to the type of variable, linearization, normalization and discretization (binning).

Data dimensionality reduction

Dimensionality reduction in the case of numerical and categorical data using various methods, PCA and multidimensional scaling. 

Data numerosity reduction

Creating a sample and balancing classes.

Introduction to cluster analysis

Distance measures, Hierarchical clustering, non-hierarchical clustering, requirements for clustering methods.

Fuzzy cluster analysis

Fuzzy values and their arithmetic, use of fuzzyin clustering methods.

Logistic regression

Basics of logistic regression and its use for data analysis.

Discriminant analysis

Advanced methods of discriminant analysis

Validation of obtained models

Model evaluation, design of control groups, stratification, etc.

Use of trees

Classification and regression trees, random forest.