Setyowati, Endah (2018) Model Hybrid Singular Spectrum Analysis dan Neural Network untuk Peramalan Nilai Pecahan Inflow dan Outflow Uang Kartal di Indonesia. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
1314100083-Undergraduate_Theses.pdf - Accepted Version Download (4MB) | Preview |
Abstract
Peredaran uang kartal di masyarakat yang sering dikenal dengan sebutan inflow dan outflow memiliki peranan yang sangat penting bagi perekonomian Indonesia. Bank Indonesia sebagai satu-satunya lembaga yang berwenang mengedarkan uang kartal kepada masyarakat harus mampu melakukan pengelolaan terhadap uang kartal melalui perencanaan mengenai kebutuhan uang dimasa mendatang dengan melakukan forecasting. Penelitian ini bertujuan untuk mengetahui pemodelan dan peramalan inflow dan outflow uang kartal setiap pecahan dengan menggunakan metode Hybrid Singular Spectrum Analysis dan Neural Network (SSA-NN). Data yang digunakan dalam penelitian ini adalah inflow dan outflow uang kartal sebanyak 14 pecahan mulai tahun 2003 hingga 2016. Analisis yang digunakan dalam penelitian ini menggunakan kajian simulasi dan kajian terapan. Hasil dari kajian simulasi menunjukkan bahwa metode SSA-NN dengan peramalan agregat cenderung lebih baik daripada individu dan pola data yang mengandung noise random lebih baik dari pada noise berpola non linier. Pada kajian terapan,hasil peramalan SSA-NN yang dibandingkan dengan ARIMAX memberikan hasil bahwa peramalan dengan menggunakan SSA-NN mampu meramalkan lebih baik sejumlah 57% pecahan inflow dan outflow uang kartal nasional. ============ The circulation of currency in the community often known as inflow and outflow has a very important role for the Indonesian economy. Bank Indonesia as the only institution that authorized to distribute currency to the public should be able to manage the currency through planning on future money needs by forecasting. This study aims to determine the modeling and forecasting of inflow and outflow of currency every fraction using Hybrid Singular Spectrum Analysis and Neural Network (SSA-NN) method. The data used in this research is the inflow and outflow currency of 14 fractions from 2003 to 2016. The analysis used in this study is using simulation studies and applied studies. The results of the simulation study indicate that the SSA-NN method with aggregate forecasting tends to have better result than the individual and the data pattern that containing random noise is better than non-linear patterned noise. In the applied studies, SSA-NN forecasting results compared to ARIMAX have resulted that forecasting using SSA-NN was able to predict better for 57% of national inflow and outflow currency.
| Item Type: | Thesis (Undergraduate) |
|---|---|
| Additional Information: | RSSt 519.536 Set m |
| Uncontrolled Keywords: | Inflow dan Outflow; Neural Network; Nonlinieritas; Pola Data; Singular Spectrum Analysis |
| Subjects: | H Social Sciences > HA Statistics H Social Sciences > HB Economic Theory > Economic forecasting--Mathematical models. H Social Sciences > HJ Public Finance Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science) |
| Divisions: | Faculty of Mathematics, Computation, and Data Science > Statistics > 49201-(S1) Undergraduate Thesis |
| Depositing User: | Endah Setyowati |
| Date Deposited: | 26 Jan 2018 02:50 |
| Last Modified: | 25 Sep 2020 03:04 |
| URI: | http://repository.its.ac.id/id/eprint/50738 |
Actions (login required)
 |
View Item |