Modeling staple food prices with autoregressive integrated moving average - Generalized autoregressive conditional heteroskedasticity (ARIMA-GARCH)

Closed

Nada Salsabila, Affiati Oktaviarina, A'Yunin Sofro, Dimas Avian Maulana, Danang Ariyanto

2025 AIP Conference Proceedings Vol. 3316 Issue 1 Conference paper Cited by 0 Quartile

Abstract

The volatility of food prices in a country can be used to evaluate its food security. Rice is a staple food for the majority of Indonesia's population. The government responds very seriously to increases in rice prices because such increases can affect the inflation rate and reduce people's welfare. The objective of this study is to model food prices using Autoregressive Integrated Moving Average - Generalized Autoregressive Conditional Heteroskedasticity (ARIMA- GARCH). This research is crucial as its findings can be used to predict food prices in East Java, which is the largest producer of staple foods in Indonesia. The method used to model food prices is ARIMA-GARCH. The best model obtained is ARIMA (3,2,1) and GARCH (3,3). The selection of the best model was based on the criteria of the smallest AIC and MAPE values, with the AIC and MAPE values of the selected model being 17.036 and 0.0218, respectively. Therefore, it can be concluded that this model can be used to forecast food prices for upcoming periods. © 2025 Author(s).

Affiliations

Mathematics Study Program, Faculty of Mathematics and Natural Science, Universitas Negeri Surabaya, East Java, Surabaya, 60231, Indonesia; Actuarial Science Study Program, Faculty of Mathematics and Natural Science, Universitas Negeri Surabaya, East Java, Surabaya, 60231, Indonesia