Universitas Airlangga Official Website

Generalized Space Time Autoregressive (GSTAR) Modeling in Predicting the Price of Bird’s Eye Chili in East Java, West Java, and Central Java

Bird’s eye chili (Capsicum frutescens L.) is a major agricultural commodity in Indonesia that contributes to the economy through high market demand and its impact on inflation. This commodity has a great opportunity to be developed, because bird’s eye chili plays a role of 80% to meet the daily needs of the community and 20% to meet the needs of the food industry. Bird’s eye chili is an essential vegetable commodity with substantial economic importance. Additionally, bird’s eye chili are demand inevitably leads to price fluctuations.

In 2022, production reached 1,544,441 tons, with East Java, Central Java, and West Java being the top producing provinces. However, price fluctuations due to production and market mismatches are a concern for farmers and policy makers. One of the causes of fluctuating prices is that farmers do not have adequate processing and preservation technology in bird’s eye chili production. To overcome the uncertain price of bird’s eye chili, appropriate analytical methods are needed to predict the price of bird’s eye chili in order to help the government to formulate policies that are right on target. This is also in line with efforts to achieve the second Sustainable Development Goals (SDG’s), namely End Hunger, Achive Food Security, and Improved Nutrition and Promote Sustainable Agriculture.

The objective of this study was to model the price dynamics of bird’s eye chili in the provinces of East Java, Central Java, and West Java, given their substantial contribution to national production. To address this, the Generalized Space Time Autoregressive (GSTAR) method was applied to model the price of bird’s eye chili from February to November 2023 using data from the National Food Agency with 8:2 ratio between training and testing data. By utilizing different weighting schemes-uniform weight, inverse distance, and cross correlation normalization, the GSTAR(21)𝐼(1) with uniform location weights performed best, showing high predictive accuracy with MAPE values of 2.021% for training data and 2.045% for test data. The model is recommended to stabilize the price of bird’s eye chili, with further validation recommended to improve reliability.

Author: Elly Pusporani, S.Si., M.Stat.

Article link: https://scholar.google.com/citations?view_op=view_citation&hl=en&user=Ssr53XAAAAAJ&citation_for_view=Ssr53XAAAAAJ:4DMP91E08xMC