Rachmawansah, Komet (2014) Average-Based Fuzzy Time Series Untuk Peramalan Kurs Valuta Asing. Sarjana thesis, Universitas Brawijaya.
Abstract
Berbagai jenis model peramalan telah banyak dikembangkan untuk meningkatkan akurasi peramalan, salah satunya adalah metode fuzzy time series. Pada penelitian ini dikemukakan tentang metode average-based fuzzy time series yang mampu menentukan panjang interval efektif, sehingga mampu memberikan hasil ramalan dengan tingkat akurasi yang baik. Metode average-based fuzzy time series ini diterapkan untuk peramalan nilai tukar mata uang yaitu USD-IDR dan EUR-USD. Hasil ramalan yang diperoleh kemudian akan dibandingkan dengan metode ARIMA. Hasil dari skripsi ini, diharapkan bisa bermanfaat untuk memperkenalkan metode average-based fuzzy time series dalam menyelesaikan masalah peramalan. Berdasarkan MSE (Mean Square Error) dan MAPE (Mean Absolute Percentage Error), untuk USD-IDR metode average-based fuzzy time series memiliki tingkat akurasi yang lebih baik dibandingkan model ARIMA(0,1,1), tetapi untuk peramalan EUR-USD model ARIMA(0,2,1) memiliki tingkat akurasi lebih baik dibandingkan metode average-based fuzzy time series akan tetapi nilai error yang dihasilkan dari kedua metode cenderung kecil. Sehingga dapat disimpulkan kedua metode layak dan baik digunakan untuk peramalan data time series khususnya pada data kurs valuta asing.
English Abstract
Various types of forecasting models have been developed to improve the accuracy of forecasting, one of which is a method of fuzzy time series. In this study the average proposed method based on fuzzy time series is capable of determining the effective length of the interval, so as to provide the forecast results with good accuracy. Average method-based fuzzy time series is applied for forecasting the exchange rate USD-USD and EUR-USD. Forecast results obtained will then be compared with ARIMA method. The results of this thesis, is expected to be useful to introduce the average method-based fuzzy time series forecasting in solving problems. Based on the MSE (Mean Square Error) and MAPE (Mean Absolute Percentage Error), for USD-IDR average -based fuzzy time series method has a better accuracy rate than the ARIMA (0,1,1), but for forecasting EUR-USD ARIMA (0,2,1) has a better accuracy rate than the average-based fuzzy time series method but the resulting error value of the two methods tend to be small. It can be concluded both feasible and well used method for forecasting time series data, especially on foreign currency exchange data.
| Item Type: | Thesis (Sarjana) |
|---|---|
| Identification Number: | SKR/MIPA/2014/286/051405285 |
| Subjects: | 500 Natural sciences and mathematics > 510 Mathematics |
| Divisions: | Fakultas Matematika dan Ilmu Pengetahuan Alam > Matematika |
| Depositing User: | Budi Wahyono Wahyono |
| Date Deposited: | 02 Sep 2014 08:23 |
| Last Modified: | 21 Oct 2021 04:50 |
| URI: | http://repository.ub.ac.id/id/eprint/153916 |
Preview |
Text
SKRIPSI_GABUNGAN.pdf Download (5MB) | Preview |
Actions (login required)
 |
View Item |