Two new time-variant methods for fuzzy time series forecasting

  • Authors:

  • Hamid Reza Kamali;Parisa Shahnazari-Shahrezaei;Hamed Kazemipoor

  • Affiliations:

  • Department of Industrial Engineering, University of Yazd, Yazd, Iran;Department of Industrial Engineering, Firoozkooh Branch, Islamic Azad University, Firoozkooh, Iran;Department of Industrial Engineering, Parand Branch, Islamic Azad University, Parand, Iran

  • Venue:
  • Journal of Intelligent & Fuzzy Systems: Applications in Engineering and Technology
  • Year:
  • 2013

Quantified Score

Hi-index 0.00

Visualization

Abstract

Nowadays, time series are widely used in forecasting. With the advent of fuzzy sets, a new gate in time series has been opened up as fuzzy time series. Basically, more information of future is being examined in fuzzy time series forecasting. Fuzzy time series methods have been extensively considered in articles and researches, especially in forecasting the historical data of statistics of Alabama University's enrollments. In this paper, two different methods are presented to accurately forecast fuzzy time series and achieve more information. To verify and validate the performance of proposed methods, four different time series including, time series with cyclic variations, a combination of linear trend and cyclic variations, exponential trend, and real values of statistics of Alabama University's enrollments are considered, too. At the end of this paper, the performance of proposed methods and existing methods in the literature are compared with each other.