Odhad modelů ratingu vícerozměrnými statistickými metodami

Abstract

The aim of diploma thesis is an estimation of the rating model for companies from the sector of the oil and gas industry through discriminant analysis and binary logistic regression. The thesis is divided into 3 main parts. The first part of the work is focused on the basic characteristics of the rating and the definition of financial ratios. In the next section are gradually explained the multivariate statistical methods that are used to estimation of rating models. In the last part are the selected statistical methods applied to specific data, which are financial ratios of the companies for the year 2010. Estimates of rating models are made in the PASW Statistics 18.0.