Skip to main content
Home

InfoTech conference

2024 International Conference on Information Technologies

The Aspects of Choosing the Time Series Length when Constructing a Fuzzy Regression Model

Liliya Anatolievna Demidova
Maksim Anatolievich Stepanov
1Department of Corporate Information Systems, Institute of Information Technologies, Russian Technological University – MIREA, Moscow
Computer Department, Faculty of Computing Engineering, Ryazan State Radio Engineering University named after V.F. Utkin, Ryazan
Russian Federation
Abstract:

The authors consider the problem of choosing the time series length when constructing a fuzzy regression model. It is proposed to perform the preliminary analysis of the time series in order to identify all the attractors, presented in them, and then apply the truncated times series, containing only a part of the consequential attractors, when constructing fuzzy regression models in order to choose the best one, in the sense of forming of an adequate corridor of fuzzy regression. An example is given that demonstrates efficacy of the proposed approach in construction of a fuzzy regression model.

Key words:
Time series
Fuzzy regression model
Corridor of fuzzy regression
Attractor