Kuddi, Bobi Frans (2016) Model Multilevel Kombinasi ARIMAX-ANFIS-GARCH Untuk Peramalan Nilai Outflow Dan Inflow Uang Kartal Di Bank Indonesia Provinsi Papua. Masters thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
1314201043-Master_Thesis.pdf - Accepted Version Download (2MB) | Preview |
Abstract
Bank Indonesia (BI) merupakan bank sentral Republik Indonesia yang
mempunyai tujuan tunggal yakni mencapai dan memelihara kestabilan nilai
rupiah. Salah satu hal yang dilakukan untuk memenuhi tujuan tersebut adalah
dengan pemantauan outflow-inflow uang kartal. Pemantauan outflow-inflow uang
kartal salah satunya dengan melakukan peramalan outflow-inflow uang kartal.
Secara umum peramalan outflow-inflow dapat dilakukan dengan pendekatan time
series, pendekatan kausal, dan gabungan antara pendekatan time series dan kausal.
Model dengan pendekatan gabungan yang banyak digunakan untuk peramalan
outflow-inflow adalah ARIMAX. Selain itu, pendekatan Autoregressive
Conditional Heteroscedasticity (ARCH) untuk model varians residual juga pernah
diaplikasikan pada peramalan outflow-inflow. Penelitian ini bertujuan mendapatkan metode terbaik untuk meramalkan outflow-inflow di BI Provinsi Papua.
Metode-metode yang akan digunakan yaitu ARIMAX dua level, ANFIS dan
gabungan ARIMAX dua level dengan ANFIS serta deteksi GARCH. Proses
GARCH untuk peramalan data outflow hanya terdapat pada model ANFIS
sendangkan pada peramalan data inflow hanya terdapat pada model ARIMAX.
Hasil yang diperoleh menunjukkan bahwa ANFIS merupakan model dengan
peramalan outflow uang kartal yang terbaik, sedangkan model yang terbaik untuk
peramalan inflow uang kartal di BI Provinsi Papua adalah metode gabungan
ARIMAX-ANFIS
====================================================================================================
Bank Indonesia (BI) is the central bank of Republic of Indonesia, which
has the sole purpose to achieve and maintain stability in the rupiah. One of the
things to do to meet these goals is monitoring the inflow-outflow of currency by
forecasting the outflow-inflow of currency. Forecasting outflow-inflow can be
generally done with time series approach, causal approach, and the combination.
Models with a combined approach that is widely used for forecasting outflowinflow is ARIMAX. In addition, the approach Autoregressive Conditional
Heteroscedasticity (ARCH) to model the residual variance has also been applied
to the forecasting of outflow-inflow. This study aims to obtain the best method for
predicting outflow-inflow in BI at Papua Province. The methods that will be used
is ARIMAX two levels, ANFIS and combined ARIMAX two levels with ANFIS
and GARCH detection. GARCH processes for forecasting the data outflow only
in ANFIS, models in forecasting inflow data is only available on models
ARIMAX. The results obtained showed that the ANFIS method is the best
forecasting model for outflow of currency, while the best model for forecasting
currency inflow in BI at Papua province is hybrid ARIMAX-ANFIS method.
| Item Type: | Thesis (Masters) |
|---|---|
| Additional Information: | RTSt 519.535 Kud m |
| Uncontrolled Keywords: | ARIMAX, ANFIS, kombinasi ARIMAX-ANFIS, GARCH, Inflow, Outflow |
| Subjects: | T Technology > T Technology (General) > T174 Technological forecasting |
| Divisions: | Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49101-(S2) Master Thesis |
| Depositing User: | Mr. Tondo Indra Nyata |
| Date Deposited: | 17 Mar 2020 02:41 |
| Last Modified: | 17 Mar 2020 02:41 |
| URI: | http://repository.its.ac.id/id/eprint/75492 |
Actions (login required)
 |
View Item |