Forecasting Volatile Fresh Chili Prices In Indonesia Using Support Vector Regression

Chili is a spicy fruit that is one of the important ingredients in many Indonesian cuisines. Most Indonesians consume fresh red chili and cayenne compared to dried chili or chili powder. In Indonesia, fresh red chili and cayenne are agricultural products with high price volatility. The volatility in chili prices has a serious effect on society. This paper aims to build accurate models to predict the prices of red chili and cayenne. It implements Support Vector Regression (SVR) with Linear, Polynomial, and RBF kernels for forecasting fresh red chili and cayenne prices from 87 cities in Indonesia. The experiments use daily time series red chili and cayenne price data from 1 January 2018 to 31 December 2023. A model is trained individually for each city and each commodity. The experimental results show that SVR Linear outperforms SVR RBF and SVR Polynomial. SVR Linear has the average of R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> scores of 0.94 for red chili prices and 0.94 for cayenne prices.