Rhinaldy, Ade (2015) Peramalan Kurs Eur/Usd Pada Pasar Forex Online Menggunakan Support Vector Regression Yang Dioptimasi Dengan Algoritma Particle Swarm Optimization. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
1211100106-Undergraduate_Thesis.pdf - Accepted Version Download (4MB) | Preview |
Abstract
Pasar forex online merupakan pasar uang yang memiliki tingkat likuiditas yang tinggi. Hal ini menyebabkan harga kurs EUR/USD pada pasar forex online bergerak fluktuatif. Para pelaku pasar forex memanfaatkan hal ini untuk mendpatkan keuntungan dengan cara melakukan posisi jual pada saat harga naik dan posisi beli pada saat harga turun. Untuk memudahkan trader dalam mengambil keputusan jual dan beli dibutuhkan peramalan data harga kurs EUR/USD. Penelitian ini bertujuan untuk memprediksi arah pergerakan harga harian perdagangan forex dengan menggunakan support vector regression (SVR). Parameter pada SVR di optimasi dengan menggunakan algoritma particle swarm optimization (PSO) sehingga didapatkan hasil peramalan yang baik. Hasil dari penelitian ini didapatkan hasil peramalan dengan nilai MAPE = 0.1895% yang dihasilkan dengan memasukkan nilai parameter optimal penalty dari SVR (C) = 102.2522 dan parameter kernel RBF ( = 1.9724.
=================================================================================================
The forex market is an online money market that has high liquidity. This led to the exchange rates EUR / USD on the online forex market moving to fluctuate. The forex market participants use this to get benefit by doing sell positions when price rises and buy positions when price fall. To facilitate traders take buy and sell decisions then forecasting of data exchange rates EUR / USD is required. Forecasting is one of mathematics branch that can be used as a tool. This study has purpose to predict the direction of daily forex price movement using support vector regression (SVR). Parameter on SVR is optimized by Particle Swarm Optimization (PSO) to obtain a good prediction. This study has results of forecasting with MAPE= 0.1895% which is produced by inserting the optimal parameter penalty from SVR (C) = 102.2522 and parameter kernel RBF (σ) = 1.9724.
| Item Type: | Thesis (Undergraduate) |
|---|---|
| Additional Information: | RSMa 658.403 55 Rhi p |
| Uncontrolled Keywords: | Peramalan, forex, Support Vector Regression, Particle Swarm Optimization |
| Subjects: | Q Science > QA Mathematics > QA278.2 Regression Analysis. Logistic regression |
| Divisions: | Faculty of Mathematics and Science > Mathematics > 44201-(S1) Undergraduate Thesis |
| Depositing User: | Yeni Anita Gonti |
| Date Deposited: | 21 Feb 2020 06:33 |
| Last Modified: | 21 Feb 2020 06:33 |
| URI: | http://repository.its.ac.id/id/eprint/75102 |
Actions (login required)
 |
View Item |