Putri, Sandria Amelia (2025) FORECASTING THE PRICE OF LARGE RED CHILI USING THE GENERALIZED SPACE-TIME AUTOREGRESSIVE – SEEMINGLY UNRELATED REGRESSION (GSTAR-SUR) MODEL IN THREE REGIONS OF EAST JAVA. Undergraduate thesis, UPN Veteran Jawa Timur.
![[img]](https://repository.upnjatim.ac.id/style/images/fileicons/text.png) |
Text (Cover)
21083010005_COVER.pdf Download (1MB) |
|
Text (Bab 1)
21083010005_BAB 1.pdf Download (287kB) |
|
Text (Bab 2)
21083010005_BAB 2.pdf Restricted to Repository staff only until 5 December 2027. Download (696kB) |
|
Text (Bab 3)
21083010005_BAB 3.pdf Restricted to Repository staff only until 5 December 2027. Download (690kB) |
|
Text (Bab 4)
21083010005_BAB 4.pdf Restricted to Repository staff only until 5 December 2027. Download (2MB) |
|
Text (Bab 5)
21083010005_BAB 5.pdf Download (226kB) |
|
Text (Daftar Pustaka)
21083010005_DAFTAR PUSTAKA.pdf Download (216kB) |
|
Text (Lampiran)
21083010005_LAMPIRAN.pdf Restricted to Repository staff only until 5 December 2027. Download (372kB) |
Abstract
Red chili is one of the main commodities contributing to volatile inflation components in Indonesia, recorded by BPS as contributing 0.06% annually to inflation in the food, beverage, and tobacco group. This is caused by fluctuations in the price of large red chilies which can be influenced by the amount of production, demand, and supply. Based on data from Siskaperbapo, the price of large red chilies in Malang Regency, Banyuwangi Regency, and Surabaya City shows a fluctuating and similar movement pattern between regions, with price increases in one region will increase prices in the other two regions and vice versa, which indicates a spatial relationship in price formation. Therefore, a forecasting model is needed that can capture spatial and temporal relationships simultaneously to support price control and food stabilization policies. This study applies the Generalized Space-Time Autoregressive – Seemingly Unrelated Regression (GSTAR-SUR) model with the Generalized Least Squares (GLS) approach to overcome correlations between residuals, and applies backward elimination as a research gap to simplify the model while retaining significant parameters. The analysis results show that the best model is GSTAR-SUR(31)-I(1) with a cross-correlation normalization weighting matrix, resulting in a MAPE of 3.25%, an Adjusted R² of 78%, an AIC of 209.31, and a BIC of 211.44, indicating a high level of accuracy. Based on the forecast for the next 14 days, the price of large red chilies is expected to decrease gradually after increasing at the end of September, with Banyuwangi Regency projected to have the highest price of IDR 46,008, followed by Surabaya City at IDR 43,988, and Malang Regency at IDR 37,479.
| Item Type: | Thesis (Undergraduate) | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Contributors: |
|
||||||||||||
| Subjects: | Q Science > QA Mathematics > QA76.6 Computer Programming | ||||||||||||
| Divisions: | Faculty of Computer Science > Departemen of Data Science | ||||||||||||
| Depositing User: | Sandria Amelia Putri | ||||||||||||
| Date Deposited: | 05 Dec 2025 08:32 | ||||||||||||
| Last Modified: | 05 Dec 2025 08:44 | ||||||||||||
| URI: | https://repository.upnjatim.ac.id/id/eprint/48057 |
Actions (login required)
 |
View Item |