Comparison of the Results of Modelling Rates of Return Depending on the Financial Situation of Companies with the Use of Real and Transformed Values of Variables

Abstract

The aim of the study is to find the relationship between rates of return on shares and indicators characterizing the financial situation of companies. Three approaches to modelling rates of return were used: classic econometric models, ordered logit models and discriminant analysis. In the first approach, the real quarterly rates of return were explained variable, and in the second and third approaches, the real rates of return were transformed into variants measured on an ordinal scale. In each approach for explanatory variables (financial indicators), time lags of up to two quarters were taken into account. Additionally, the explanatory variables were transformed—the nominants were transformed into stimulants or destimulants. The study used quarterly data from 2014 to 2017 on financial ratios for Machinery sector companies listed on the Warsaw Stock Exchange from the Notoria Serwis database and data on prices from Bank Ochrony Środowiska. As a result of research conducted with the use of particular models, the financial indicators which had the greatest impact on the level of return rates were identified. Knowledge of these indicators may help investors to make decisions on the selection of companies for the investment portfolio.