Syllabus
1) Basics of linear regression, logistic regression, lasso regression, principles of hypotheses testing, likelihood ratio tests, stepwise algorithms
2) Basics of multidimensional statistics - principle component analysis, factor analysis, cluster analysis
3) Discrimination measures - Kolmogorov-Smirnov, Gini coefficient, Somer’s d
4) Back test principles, cross validation and bootstrapping
5) Regression trees, random forests
6) Gradient boosting
7) Bayes networks, neural networks
8) Linear optimization, Support vector machine Labs: Programming in R, practical work with data
Annotation
The lecture covers standard methods of data analysis, including modern trends of big data analysis using machine learning. Modelling over real data in the R environment.