Time Series and Its Characteristics Variability Analysis Using the Forecasting Model Based on the Basis of Interval Discrete Fuzzy Sets of the Second Type

Abstract

The approach to the problem of analyzing the variability of the time series and its characteristics in the context of developing a decision on the advisability of implementing the short-term forecasting procedure for one step forward has been considered. The Hurst exponent and fractal dimension are used as the characteristics of the time series. It is proposed to use the forecasting model on the basis of interval discrete fuzzy sets of the second type with the aim of analyzing the parameters of the generated “footprint of uncertainty” for the initial time series and the time series describing one or another of its characteristics, and deciding on the alleged success or failure of the short-term forecasting procedure for one step forward. The examples that confirm the effectiveness of the proposed approach to the problem of analyzing the variability of the time series have been given.