Multiset Based Forecasting Model for Fuzzy Time Series

Abstract

Since the pioneering work of Song and Chissom (1993a, b) on fuzzy time series to model and forecast processes whose values are described by linguistic values, a number of techniques have been proposed by researchers for forecasting. In most of the realistic situation the duplicates of data are significant. This paper presents a new fuzzy time series method, which employs multiset theory. The historical data of daily average temperature in Taipei, Taiwan (central weather bureau 1996) are adopted to illustrate the forecasting process of the proposed method.

Best practice as actual and relative benchmark to inefficient units: Multiset DEA analysis

“…According to certain authors, the problem is solved by consecutive decision-making, where a new multiset function of loss is proposed as a parameter of predictive policy (Welleck et al, 2017). According to others, the multiset approach is used to predict the average daily temperature, as shown by the Taipei example in Taiwan (Vamitha & Rajaram, 2015). In our paper, the Multiset DEA analysis of units is used for predicting inefficient results, which meant increasing the set of decision-making units by adding a new set.…”

Aggregation and definition of an algebraic framework for fuzzy time series: An application in the supply-demand domain

“…The following is a series of articles that through the application of fuzzy techniques allow the resolution of problems with such characteristic. As topics, there can be found enrollment [13,14,15], temperature [16,17,18], reactors [19], the concentration of pollutant gases [20,21], tourism [22,23,24] and aspects related to COVID-19 disease [25,26], among others. But undoubtedly, a large majority of applications are focused on forecasting.…”

Fuzzy Time Series Forecasting: Chen, Markov Chain and Cheng Models

“…An advantage of fuzzy time series is that it relies on simple and uncomplicated calculations compared to genetic algorithms, including Neural network models, and also can more efficiently use historical data since they can associate trend or cyclic components in fuzzy logical relationships. However, the neural network models are better at handling non-linear problems, but they are less effective due to lengthy training periods and their predicted values are less precise as a result of their inability to handle non-stationary data (Li and Cheng 2007;Vamitha et al, 2012). Several methods of fuzzy time series among them, the Chen method, the Cheng method, the Markov chain method, and others, each of these methods has a different methodology for forecasting (Rachim et al, 2020).…”