Wicaksono, Benedictus Satriyo Karuna and Dr. Dra. Umu Sa‘adah, M.Si. (2024) Analisis Risiko Kurs Transaksi Bank Indonesia Menggunakan Metode Value at Risk Dengan Pendekatan Hybrid ARIMA-GARCH. Sarjana thesis, Universitas Brawijaya.
Abstract
Penelitian ini membahas manajemen risiko dalam pertukaran return kurs jual EUR (Euro) dengan IDR (Indonesian Rupiah) Bank Indonesia menggunakan metode peramalan Autoregressive Integrated Moving Average (ARIMA) dan Generalized Autoregressive Conditional Heteroscedasticity (GARCH). Analisis dilakukan dengan memanfaatkan Hybrid ARIMA-GARCH untuk mengestimasi risiko Value at Risk (VaR) dari data historis return kurs jual EUR/IDR Bank Indonesia. Metode ini dilakukan untuk menangani pola volatilitas dalam data. Hasil analisis menunjukkan model terbaik dari data training untuk model ARIMA adalah ARIMA([5], 0, [9][16]). Namun berdasarkan pemeriksaan diagnostik didapatkan bahwa residual tidak memenuhi asumsi white noise. Selanjutnya dilakukan pemeriksaan efek ARCH terhadap residual. Karena terdapat efek ARCH terhadap residual, maka dilakukan pemodelan lebih lanjut menggunakan GARCH. Hasil pemodelan didapatkan model terbaik untuk GARCH adalah GARCH(1,1). Dengan demikian diperoleh model hybrid adalah ARIMA([5],0,[9][16])-GARCH(1,1). Evaluasi performa dengan Mean Square Percentage Error (MAPE) menunjukkan bahwa model Hybrid ARIMA-GARCH memiliki tingkat akurasi sangat akurat pada data training dan testing, yaitu berturut-turut 3.939277% dan 1.470776%. Estimasi VaR dari model Hybrid ARIMA-GARCH sebesar 0.023021%. Melalui Backtesting, nilai VaR model mampu memprediksi secara akurat dengan jumlah pelanggaran sebanyak tiga. Penilaian ini menegaskan bahwa kemampuan model Hybrid ARIMA-GARCH sangat baik dalam meramalkan kerugian terburuk bagi trader selama melakukan perdagangan kurs jual EUR/IDR
English Abstract
This study discusses risk management in the exchange of Euro (EUR) selling rates with Indonesian Rupiah (IDR) at Bank Indonesia using the Autoregressive Integrated Moving Average (ARIMA) and Generalized Autoregressive Conditional Heteroscedasticity (GARCH) forecasting methods. The analysis utilizes a Hybrid ARIMA-GARCH approach to estimate the Value at Risk (VaR) of historical return data for EUR/IDR selling rates at Bank Indonesia, addressing volatility patterns in the data. The analysis results indicate that the best model from the training data for the ARIMA model is ARIMA([5], 0, [9][16]). However, diagnostic checks reveal that the residuals do not meet the white noise assumption. Subsequently, ARCH effects on residuals are examined. Due to the presence of ARCH effects on residuals, further modeling is conducted using GARCH. The modeling results show that the best model for GARCH is GARCH(1,1). Thus, the hybrid model obtained is ARIMA([5],0,[9][16])-GARCH(1,1). Performance evaluation using Mean Square Percentage Error (MAPE) indicates that the Hybrid ARIMA-GARCH model exhibits highly accurate accuracy levels on training and testing data, respectively 3.939277% and 1.470776%. The VaR estimation from the Hybrid ARIMA-GARCH model is 0.023021%. Through backtesting, the VaR model accurately predicts violations with a total of three. This assessment confirms that the Hybrid ARIMA-GARCH model’s ability is excellent in forecasting worst-case losses for traders during EUR/IDR selling rate trading.
| Item Type: | Thesis (Sarjana) |
|---|---|
| Identification Number: | 0524090090 |
| Uncontrolled Keywords: | ARIMA, Backtesting, GARCH, peramalan, Value at Risk |
| Divisions: | Fakultas Matematika dan Ilmu Pengetahuan Alam > Matematika |
| Depositing User: | Unnamed user with username nova |
| Date Deposited: | 27 Jun 2024 08:35 |
| Last Modified: | 27 Jun 2024 08:35 |
| URI: | http://repository.ub.ac.id/id/eprint/220461 |
![[thumbnail of DALAM MASA EMBARGO]](http://repository.ub.ac.id/style/images/fileicons/text.png) |
Text (DALAM MASA EMBARGO)
BENEDICTUS SATRIYO KARUNA WICAKSONO.pdf Restricted to Registered users only Download (3MB) |
Actions (login required)
 |
View Item |