Fuzzy Model for Estimation and Prediction of Stock Market System
Alexander N. Kozlovsky, Alexey O. Nedosekin, Mikhail S. Kokorin · 2020
Purpose: To propose a method for evaluating and forecasting stock market tendencies based on a fuzzy function, which is a fuzzy regression of the price-to-earnings ratio (PE). Method: In the value estimation model, the PE factor is interpreted as fuzzy parabolic regression, the parameters of which are functions of time. Thus, PE is presented as a dynamic K-lens, and the predicted business value is presented as an interval fuzzy function of time. Result: Based on data on 250 international companies with the largest capitalization for the period from 2016 to 2018, it is possible to construct a parabolic regression of the PE factor. All companies that fall outside the scope of the constructed R-lens in terms of the “Price-to-book” (PB) indicator provides the market player with the opportunity to earn income by acquiring derivative financial instruments of an appropriate orientation. Conclusions: If there is a reliable PE forecast for the market as a whole, then it can be transformed into a forecast for parameters of the corresponding regression. This gives reason to predict the capitalization of a particular company.