A heuristic time-invariant model for fuzzy time series forecasting

Bai, Enjian; Wong, Wai Keung; Chu, Wai Ching; Xia, Min; Pan, Feng

doi:10.1016/j.eswa.2010.08.059

Cited by 36 publications

(13 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Selain tidak memerlukan data latih yang banyak fuzzy time series juga cocok digunakan untuk peramalan jangka panjang maupun peramalan jangka pendek dengan hasil akurasi prediksi yang akurat (Bai et al, 2011).…”

Section: Kerangka Teoriunclassified

“…Fuzzy relationship dan nilainilai yang diprediksi ditentukan oleh sepuluh aturan heuristik. Dalam model yang diusulkan, orde maksimum w fuzzy relationship yang akan dibentuk dalam tahap pelatihan diatur untuk mengurangi kompleksitas perhitungan (Bai et al, 2011 (Chen, 1996;Huarng, 2001). …”

Section: Permodelan Fuzzy Time Series Heuristic Time Invariantunclassified

“…Perhitungan For1 dan For2 dilakukan dengan mengikuti aturan (rule) berikut (Bai et al, 2011): Aturan 1 : Jika orde ke k fuzzy relationship group adalah k ke satu, misalnya : , ( −1) , … , 1 → Nilai keanggotaan maksimum pada Aj terjadi di interval uj, maka prediksinya mj. = (4) Aturan 2 : Jika orde ke k fuzzy relationship group adalah k ke banyak, misalnya, , ( −1) , … , 1 → 1 , 2 , nilai keanggotaan maksimum pada Ajt terjadi di interval ujt, maka nilai prediksinya adalah rata-rata dari mjt, nilai tengah pada ujt, dimana t = 1, 2, … , p.…”

Section: Permodelan Fuzzy Time Series Heuristic Time Invariantunclassified

See 2 more Smart Citations

Model Heuristic Time Invariant Fuzzy Time Series dan Regresi Untuk Prediksi Laba dan Analisis Variabel yang Mempengaruhi

Admirani

Gernowo

Suryono

2016

J. Sistem Info. Bisnis

View full text Add to dashboard Cite

Model of prediction with fuzzy time series method has ability to capture the pattern of past data to predict the fu ture of data does not need a complicated system, making it easier to use. The research aims to built prediction system using model of heuristic time invariant fuzzy time series and multiple linear regression to predict profit and analysis of variables that affect profit. Profit forecasting aims to determine the company's prospects in the future in order to remain exist in doing its business. The variables that use in the modelling are profit as the dependent variable, and sales, cost of goods sold, general and administrative expenses, selling and marketing expenses and interest income as the indepent variables. Profit forecasting modelling begins by defining universe of discourse and interval actual data of profit, then determine fuzzy set and actual data fuzzified. Furthermore, fuzzy logical relationship and fuzzy logical relationships group to fuzzified data. The prediction process consist of two prediction phase there are training phase aimed to determine trend predictor and testing phase to determine prediction results. By using 24 profit data samples resulted prediction error by using Mean Absolute Percentage Error is 11,64% and added 13 data for testing obtained prediction error is 22,27%. In analysis of variables that affect profit is known that sales variable most effect on profit than other variables with a regression coefficient 0.976.Keywords : Profit forecast; heuristic time invariant fuzzy time series; multiple linear regression AbstrakModel prediksi dengan metode fuzzy time series mempunyai kemampuan untuk dapat menangkap pola data yang telah lalu untuk memprediksi data yang akan datang tidak membutuhkan sistem yang rumit, sehingga lebih mudah untuk digunakan. Penelitian ini bertujuan membuat system predisi menggunakan model heuristic time invariant fuzzy time series dan regresi linear berganda untuk prediksi laba dan analisis variabel yang mempengaruhinya. Prediksi laba bertujuan untuk mengetahui prospek perusahaan di masa mendatang supaya tetap eksis menjalankan usahanya. Variabel yang digunakan adalah laba, penjualan, beban pokok penjualan, beban umum dan administrasi, beban penjualan dan pemasaran dan penghasilan bunga yang dimasukkan langsung kedalam sistem. Pembuatan model prediksi laba dimulai dengan menentukan semesta pembicaraan dan interval dari data aktual laba, kemudian menentukan himpunan fuzzy dan fuzzifikasi data aktual. Selanjutnya dilakukan hubungan logika fuzzy dan pengelompokan hubungan logika fuzzy terhadap data hasil fuzzifikasi. Proses prediksi terbagi atas dua tahapan yaitu tahap training yang bertujuan untuk menentukan prediktor tren, dan tahap testing untuk menentukan hasil prediksi. Dengan input data menggunakan 24 data sampel laba diperoleh error prediksi menggunakan Mean Absolute Percentage Error sebesar 11,64%, den dengan 13 data laba untuk testing diperoleh error prediksi sebesar 22,27%. Dalam analisis variabel yang mempengaruhi laba diketahui bah...

show abstract

Section: Kerangka Teoriunclassified

Section: Permodelan Fuzzy Time Series Heuristic Time Invariantunclassified

See 1 more Smart Citation

Model Heuristic Time Invariant Fuzzy Time Series dan Regresi Untuk Prediksi Laba dan Analisis Variabel yang Mempengaruhi

Admirani

Gernowo

Suryono

2016

J. Sistem Info. Bisnis

View full text Add to dashboard Cite

show abstract

“…A heuristic time-invariant series model, which fully utilized the trend predicator to help forecasting and to ultimately achieve higher accuracy results but disregarded the effect of the corresponding external time series, was discussed. 35 …”

Section: Literature Reviewmentioning

confidence: 99%

Heuristic Bivariate Forecasting Model of Multi-Attribute Fuzzy Time Series Based on Fuzzy Clustering

Nan

Zhou

Kou

et al. 2012

Int. J. Info. Tech. Dec. Mak.

View full text Add to dashboard Cite

Fuzzy time series has been applied to forecast various domain problems because of its capability to deal with vagueness and incompleteness inherent in data. However, most existing fuzzy time series models cannot cope with multi-attribute time series and remain too subjective in the partition of the universe of discourse. Moreover, these models do not consider the trend factor and the corresponding external time series, which are highly relevant to target series. In the current paper, a heuristic bivariate model is proposed to improve forecasting accuracy, and the proposed model applies fuzzy c-means clustering algorithm to process multi-attribute fuzzy time series and to partition the universe of discourse. Meanwhile, the trend predictors are extracted in the training phase and utilized to select the order of fuzzy relations in the testing phase. Finally, the proper full use of the external series to assist forecasting is discussed. The performance of the proposed model is tested using actual time series including the enrollments at the University of Alabama, the Taiwan Stock Exchange Capitalization Weighted Stock Index (TAIEX) and a sensor dataset. The experimental results show that the proposed model can be utilized for multi-attribute time series and significantly improves the average MAER to 1.19% when compared with other forecasting models.

show abstract

“…Enrolment data is one of these data types that has quite a wide range of usage areas in literature from the beginning of studies [4, 12, 19, 23, 24, 27, 29, 30, 32-34, 37, 38, 44, 51, 53, 54, 55-58, 63, 65, 67, 72, 75, 78, 80, 82-84, 87, 89, 95, 97]. In addition Taiwan Futures Exchange (TAIFEX) [4,12,32,58,59,65,68,91] Taiwan Stock Exchange Capitalization Weighted Stock Index (TAIEX) [17-19, 25-28, 33, 36, 38-40, 42, 57, 59-61, 73, 93-95, 99-101] and Index 100 in stock and bonds exchange market of Istanbul (IEX) [4] have been also widely utilized for getting accurate prediction in FTS literature. And also, for temperature prediction, various FTS forecasting methods were used in the same area [25, 31, 32, 42, 68-70, 73, 80, 91].…”

Section: Literature Reviewmentioning

confidence: 99%

A combined robust fuzzy time series method for prediction of time series

Yolcu

Lam

2017

Neurocomputing

View full text Add to dashboard Cite

In case of outlier(s) it is inevitable that the performance of the fuzzy time series prediction methods is influenced adversely. Therefore, current prediction methods will not be able to provide satisfactory accuracy rates for defuzzified outputs (predictions) when the data has outlier(s). In this study, not only to be able to sort out this problem but also to be able to improve the forecasting accuracy, we propose a combined robust approach for fuzzy time series by assessing how the prediction performance of the methods will be affected from the

show abstract

A heuristic time-invariant model for fuzzy time series forecasting

Cited by 36 publications

References 25 publications

Model Heuristic Time Invariant Fuzzy Time Series dan Regresi Untuk Prediksi Laba dan Analisis Variabel yang Mempengaruhi

Model Heuristic Time Invariant Fuzzy Time Series dan Regresi Untuk Prediksi Laba dan Analisis Variabel yang Mempengaruhi

Heuristic Bivariate Forecasting Model of Multi-Attribute Fuzzy Time Series Based on Fuzzy Clustering

A combined robust fuzzy time series method for prediction of time series

Contact Info

Product

Resources

About