An improved fuzzy time series forecasting method using trapezoidal fuzzy numbers

Abstract

One of the major drawbacks of the existing fuzzy time series forecasting models is the fact that they only provide a single-point forecasted value just like the output of the traditional time series methods. Hence, they cannot provide a decision analyst more useful information. The aim of this present research is to design an improved fuzzy time series forecasting method in which the forecasted value will be a trapezoidal fuzzy number instead of a single-point value. Furthermore, the proposed method may also increase the forecasting accuracy. Two numerical data sets were used to illustrate the proposed method and compare the forecasting accuracy with three fuzzy time series methods. The results of the comparison indicate that the proposed method can generate forecasting values that are more accurate.

Comparative analysis of forecasting portfolio returns using Soft Computing technologies

“…On the other hand, Liu [27] FTS forecasting method also maintains the partitioning of the universe of discourse suggested by Huarng (the average-based length method), but calculates the forecasted outputs as trapezoidal fuzzy numbers, using the endpoints of the intervals u i .…”

“…We use this information in the so-called Chen-weighted method, which takes into account those frequencies, and provides as pointwise fuzzy forecast a weighted average of the linguistic values of the linguistic variable 'return', with weights proportional to the relative frequencies of the elements in the corresponding fuzzy logical relationship group. In addition, following [27] we define the membership function of the seven linguistic values with trapezoidal numbers. This number is a weighted average of the trapezoidal numbers representing the seven linguistic values that compose the universe of discourse; the weights are given by the vector (0.0, 0.0, 0.11, 0.44, 0.40, 0.02, 0.02) whose components are the relative frequencies of the fuzzy logical relationship of A 5 (the linguistic value of the last observed value in the training set r (150,1) = 1.39) with the other linguistic values into its relationship group.…”

A Novel Forecasting Method Based on F-Transform and Fuzzy Time Series

“…Table 7 shows the forecasted values from 1993 to 2006. Taiwan Table 8 shows the number of patents granted in Taiwan from 1980 to 2000 introduced by Liu [21]. The data set is transformed to be invariant time series by using first-order time difference because it is increasing.…”

“…In Sect. 4, the experimental applications with enrollment data [32] and patents granted data of Taiwan [21] are provided. Section 5 concludes this study.…”

Fuzzy Regression Model Using Trapezoidal Fuzzy Numbers for Re-auction Data

“…Fuzzy regression analysis using triangular fuzzy numbers and trapezoidal fuzzy number have been studied in many works [9][10][11][12][13][14][15]. The theoretical studeis regarding fuzzy regression model have been investigated in [16][17][18].…”