Model VARIMA Untuk Peramalan Rata-rata Bulanan Harga Cabai Rawit Merah, Merah Keriting, Dan Merah Besar Di Jawa Timur

Pasaribu, Gita Lestari (2026) Model VARIMA Untuk Peramalan Rata-rata Bulanan Harga Cabai Rawit Merah, Merah Keriting, Dan Merah Besar Di Jawa Timur. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5002221067-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (5MB) | Request a copy

Abstract

Cabai termasuk komoditas hortikultura dengan harga yang sering berfluktuasi sehingga peramalan diperlukan untuk memperkirakan pergerakan harganya pada periode mendatang. Di Jawa Timur, jenis cabai yang banyak dikonsumsi masyarakat meliputi cabai rawit merah, cabai merah keriting, dan cabai merah besar. Ketiga jenis cabai tersebut memiliki pola pergerakan harga yang saling berkaitan sehingga ketiganya dapat diramalkan secara bersama menggunakan Vector Autoregressive Integrated Moving Average (VARIMA). Penelitian ini menggunakan data harga cabai harian yang diolah menjadi rata-rata bulanan yang dibagi menjadi data training dan testing. Model VARIMA dibangun melalui tahap identifikasi model, estimasi parameter menggunakan metode Conditional Maximum Likelihood, uji siginfikansi parameter, uji diagnostik residual, dan peramalan. Hasil identifikasi menunjukkan bahwa model yang terpilih adalah VARMA(4,0) atau VAR(4). Kinerja model tersebut kemudian dievaluasi melalui nilai MAPE, yaitu sebesar 30,32% untuk cabai rawit merah, 19,86% untuk cabai merah keriting, dan 21,68% untuk cabai merah besar, dengan rata-rata MAPE sebesar 23,95%. Nilai tersebut menunjukkan bahwa model VAR(4) memiliki kemampuan peramalan yang cukup memadai sehingga digunakan untuk meramalkan rata-rata bulanan harga ketiga jenis cabai pada periode April hingga Desember 2026. Hasil penelitian ini diharapkan dapat menjadi informasi pendukung bagi pihak yang membutuhkan.
===============================================================================================================================
Chili is a horticultural commodity whose prices frequently fluctuate, making forecasting essential for estimating future price movements. In East Java, the chili varieties most commonly consumed are red cayenne chili, curly red chili, and large red chili. Since the price movements of these three varieties are interrelated, they can be forecast simultaneously using the Vector Autoregressive Integrated Moving Average (VARIMA) method. This study uses daily chili price data aggregated into monthly averages and divided into training and testing datasets. The VARIMA model is developed through model identification, parameter estimation using the Conditional Maximum Likelihood method, parameter significance testing, residual diagnostic testing, and forecasting. The identification results indicate that the selected model is VARMA(4,0), which is equivalent to VAR(4). The model's forecasting performance is evaluated using the Mean Absolute Percentage Error (MAPE), resulting in values of 30.32% for red cayenne chili, 19.86% for curly red chili, and 21.68% for large red chili, with an average MAPE of 23.95%. These results indicate that the VAR(4) model provides reasonably accurate forecasting performance and can therefore be used to forecast the average monthly prices of the three chili varieties from April to December 2026. The findings of this study are expected to provide valuable supporting information for policymakers and other relevant stakeholders.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Harga Cabai, Jawa Timur, Peramalan, Vector Autoregressive Integrated Moving Average (VARIMA), Chili Prices, East Java, Forecasting, Vector Autoregressive Integrated Moving Average (VARIMA)
Subjects:	Q Science > QA Mathematics Q Science > QA Mathematics > QA276 Mathematical statistics. Time-series analysis. Failure time data analysis. Survival analysis (Biometry) Q Science > QA Mathematics > QA401 Mathematical models.
Divisions:	Faculty of Mathematics and Science > Mathematics > 44201-(S1) Undergraduate Thesis
Depositing User:	Gita Lestari Pasaribu
Date Deposited:	28 Jul 2026 01:19
Last Modified:	28 Jul 2026 01:19
URI:	http://repository.its.ac.id/id/eprint/138100

Actions (login required)

View Item