Hidayatullah, M. Maulana Sholihin (2017) Peramalan Kenaikan Indeks Harga Konsumen/Inflasi Kota Malang Menggunakan Metode Support Vector Regression (SVR) Dengan Chaotic Genetic Algorithm-Simulated Annealing (CGASA). Sarjana thesis, Universitas Brawijaya.
Abstract
Peramalan inflasi adalah hal yang rumit. Tingkat inflasi yang dihitung berdasarkan kenaikan indeks harga konsumen (IHK) dipengaruhi berbagai faktor mulai dari gejolak harga berbagai jenis barang yang tidak menentu, nilai tukar rupiah, tingkat inflasi dunia, kebijakan pemerintah, gejolak suplai barang dan permintaan masyarakat. Hibridasi algoritma support (SVR) dengan chaotic sequence dan algoritma genetika telah sukses diaplikasikan untuk meningkatkan akurasi peramalan dalam berbagai bidang. Tetapi masih belum banyak diekplorasi penggunaan algoritma ini dalam bidang ekonomi pasar yaitu peramalan inflasi. Jurnal ini akan menganalisis potensial dari algoritma hibridasi yaitu chaotic genetic algorithm-simulated annealing algorithm (CGASA) dengan model SVR untuk meningkatkan performa akurasi peramalan. Dengan tingkat keacakan yang kacau dari chaotic sequence akan mampu menghindarkan premature local optimum dan korvengensi dini, terlebih dengan adanya algoritma simulated annealing yang meningkatkan wilayah pencarian solusi. Hasil uji peramalan pada penelitian ini menunjukkan keakuratan yang lebih baik dibandingkan penelitian sebelumnya yang telah dikaji yaitu Metode ensembel gabungan antara algoritma autoregressive integrated moving average (ARIMA) dan jaringan syaraf tiruan (ANN).
English Abstract
Inflation forecasting is complicated. Inflation rate calculated based on the rise in the consumer price index (CPI) is influenced by various factors ranging from volatile prices of various types of goods, rupiah exchange rate, world inflation rate, government policy, fluctuations in the supply of goods and demand. Hybridation algorithm support (SVR) with chaotic sequences and genetic algorithms has been successfully applied to improve the accuracy of forecasting in various fields. But it has not been explored the usablity of this algorithm in the field of market economy which is forecasting inflation. This journal will analyze the potential of hybridization algorithm that which is chaotic genetic algorithm-simulated annealing algorithm (CGASA) with SVR model to improve the performance of forecasting accuracy. With the chaotic sequence of chaotic sequences, it will be able to avoid premature local optimum and early corvengensi, especially with the simulated annealing algorithm that increases the search area of the solution. The results of the forecasting test in this study show better accuracy than the previous research which has been studied is the combined ensemble method between autoregressive integrated moving average (ARIMA) and artificial neural network (ANN) algorithm.
| Item Type: | Thesis (Sarjana) |
|---|---|
| Identification Number: | SKR/FTIK/2017/496/051707818 |
| Uncontrolled Keywords: | Peramalan Inflasi, Indeks Harga Konsumen, Support Vector Regression (SVR), Chaotic Genetic Algorithm-Simulated Annealing (CGASA) |
| Subjects: | 000 Computer science, information and general works > 003 Systems > 003.2 Forecasting and forecasts |
| Divisions: | Fakultas Ilmu Komputer > Teknik Informatika |
| Depositing User: | Yusuf Dwi N. |
| Date Deposited: | 04 Sep 2017 05:25 |
| Last Modified: | 18 Sep 2020 02:42 |
| URI: | http://repository.ub.ac.id/id/eprint/2023 |
Preview |
Text
M. Maulana Sholihin Hidayatullah.pdf Download (2MB) | Preview |
Actions (login required)
 |
View Item |