Використання нечіткої міри для подолання невизначеності довгострокових прогнозів на основі екстраполяцій

The method to overcome the uncertainty of long-term predictions, performed in the form of extrapolations based on the observational data, which are represented by the variation series or uniform, monotonous time series, with using the fuzzy measure is proposed. For the approximating models in the fo...

Full description

Saved in:
Bibliographic Details
Date:2013
Main Authors: Stefanyshyna-Gavryliuk, Yu. D., Stefanyshyn, D. V.
Format: Article
Language:Ukrainian
Published: The National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute" 2013
Online Access:http://journal.iasa.kpi.ua/article/view/33948
Tags: Add Tag
No Tags, Be the first to tag this record!
Journal Title:System research and information technologies

Institution

System research and information technologies
Description
Summary:The method to overcome the uncertainty of long-term predictions, performed in the form of extrapolations based on the observational data, which are represented by the variation series or uniform, monotonous time series, with using the fuzzy measure is proposed. For the approximating models in the form of extrapolations in case of variation series the probability distribution functions are used, and in case of time-series the trends are used. Different variants of approximating models are considered as expert evaluations. The results of forecasting are processed using methods of the theory of fuzzy sets and fuzzy logic. It is shown that in order to overcome linguistic uncertainty of forecast results, obtained using different extrapolation models, the fuzzy variables the functions of which are set by means of the values of the reliability of hypotheses about the probability distribution functions and the coefficients of determination for trends, may be used.