Deterministic vector long-term forecasting for fuzzy time series

Sheng Tun Li, Shu Ching Kuo, Yi Chung Cheng, Chih Chuan Chen

Research output: Contribution to journalArticle

36 Citations (Scopus)

Abstract

In the last decade, fuzzy time series have received more attention due their ability to deal with the vagueness and incompleteness inherent in time series data. Although various improvements, such as high-order models, have been developed to enhance the forecasting performance of fuzzy time series, their forecasting capability is mostly limited to short-term time spans and the forecasting of a single future value in one step. This paper presents a new method to overcome this shortcoming, called deterministic vector long-term forecasting (DVL). The proposed method, built on the basis of our previous deterministic forecasting method that does not require the overhead of determining the order number, as in other high-order models, utilizes a vector quantization technique to support forecasting if there are no matching historical patterns, which is usually the case with long-term forecasting. The vector forecasting method is further realized by seamlessly integrating it with the sliding window scheme. Finally, the forecasting effectiveness and stability of DVL are validated and compared by performing Monte Carlo simulations on real-world data sets.

Original languageEnglish
Pages (from-to)1852-1870
Number of pages19
JournalFuzzy Sets and Systems
Volume161
Issue number13
DOIs
Publication statusPublished - 2010 Jul 1

Fingerprint

Fuzzy Time Series
Forecasting
Time series
Higher Order
Time Series Forecasting
Vagueness
Vector Quantization
Incompleteness
Sliding Window
Pattern matching
Time Series Data
Vector quantization
Monte Carlo Simulation

All Science Journal Classification (ASJC) codes

  • Logic
  • Artificial Intelligence

Cite this

Li, Sheng Tun ; Kuo, Shu Ching ; Cheng, Yi Chung ; Chen, Chih Chuan. / Deterministic vector long-term forecasting for fuzzy time series. In: Fuzzy Sets and Systems. 2010 ; Vol. 161, No. 13. pp. 1852-1870.
@article{3f793a1e9917405ab7e56eafd45ee0ae,
title = "Deterministic vector long-term forecasting for fuzzy time series",
abstract = "In the last decade, fuzzy time series have received more attention due their ability to deal with the vagueness and incompleteness inherent in time series data. Although various improvements, such as high-order models, have been developed to enhance the forecasting performance of fuzzy time series, their forecasting capability is mostly limited to short-term time spans and the forecasting of a single future value in one step. This paper presents a new method to overcome this shortcoming, called deterministic vector long-term forecasting (DVL). The proposed method, built on the basis of our previous deterministic forecasting method that does not require the overhead of determining the order number, as in other high-order models, utilizes a vector quantization technique to support forecasting if there are no matching historical patterns, which is usually the case with long-term forecasting. The vector forecasting method is further realized by seamlessly integrating it with the sliding window scheme. Finally, the forecasting effectiveness and stability of DVL are validated and compared by performing Monte Carlo simulations on real-world data sets.",
author = "Li, {Sheng Tun} and Kuo, {Shu Ching} and Cheng, {Yi Chung} and Chen, {Chih Chuan}",
year = "2010",
month = "7",
day = "1",
doi = "10.1016/j.fss.2009.10.028",
language = "English",
volume = "161",
pages = "1852--1870",
journal = "Fuzzy Sets and Systems",
issn = "0165-0114",
publisher = "Elsevier",
number = "13",

}

Deterministic vector long-term forecasting for fuzzy time series. / Li, Sheng Tun; Kuo, Shu Ching; Cheng, Yi Chung; Chen, Chih Chuan.

In: Fuzzy Sets and Systems, Vol. 161, No. 13, 01.07.2010, p. 1852-1870.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Deterministic vector long-term forecasting for fuzzy time series

AU - Li, Sheng Tun

AU - Kuo, Shu Ching

AU - Cheng, Yi Chung

AU - Chen, Chih Chuan

PY - 2010/7/1

Y1 - 2010/7/1

N2 - In the last decade, fuzzy time series have received more attention due their ability to deal with the vagueness and incompleteness inherent in time series data. Although various improvements, such as high-order models, have been developed to enhance the forecasting performance of fuzzy time series, their forecasting capability is mostly limited to short-term time spans and the forecasting of a single future value in one step. This paper presents a new method to overcome this shortcoming, called deterministic vector long-term forecasting (DVL). The proposed method, built on the basis of our previous deterministic forecasting method that does not require the overhead of determining the order number, as in other high-order models, utilizes a vector quantization technique to support forecasting if there are no matching historical patterns, which is usually the case with long-term forecasting. The vector forecasting method is further realized by seamlessly integrating it with the sliding window scheme. Finally, the forecasting effectiveness and stability of DVL are validated and compared by performing Monte Carlo simulations on real-world data sets.

AB - In the last decade, fuzzy time series have received more attention due their ability to deal with the vagueness and incompleteness inherent in time series data. Although various improvements, such as high-order models, have been developed to enhance the forecasting performance of fuzzy time series, their forecasting capability is mostly limited to short-term time spans and the forecasting of a single future value in one step. This paper presents a new method to overcome this shortcoming, called deterministic vector long-term forecasting (DVL). The proposed method, built on the basis of our previous deterministic forecasting method that does not require the overhead of determining the order number, as in other high-order models, utilizes a vector quantization technique to support forecasting if there are no matching historical patterns, which is usually the case with long-term forecasting. The vector forecasting method is further realized by seamlessly integrating it with the sliding window scheme. Finally, the forecasting effectiveness and stability of DVL are validated and compared by performing Monte Carlo simulations on real-world data sets.

UR - http://www.scopus.com/inward/record.url?scp=77950628869&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77950628869&partnerID=8YFLogxK

U2 - 10.1016/j.fss.2009.10.028

DO - 10.1016/j.fss.2009.10.028

M3 - Article

AN - SCOPUS:77950628869

VL - 161

SP - 1852

EP - 1870

JO - Fuzzy Sets and Systems

JF - Fuzzy Sets and Systems

SN - 0165-0114

IS - 13

ER -