Modelling and forecasting of onion prices in Belgaum market of Karnataka

Abstract

Onions are considered as one of the most sensitive commodities due to their susceptibility to sudden price fluctuations. Therefore, modelling and forecasting commodity prices, particularly for volatile vegetables like onions, are crucial. In our current study, we applied various statistical techniques, including Single Exponential Smoothing, Double Exponential Smoothing, Triple Exponential Smoothing, ARIMA, SARIMA, BATS, and TBATS models, to analyse onion prices from January 2010 to December 2023 in the Belgaum market of Karnataka state. The models were built using an 80-20 split, with 80% of the observations used to the training dataset and the remaining 20% to the testing dataset for model validation. Among the seven models assessed, the BATS model consistently outperformed the others, showcasing lower Root Mean Square Error (RMSE) and Mean Absolute Percentage Error (MAPE) values. Additionally, the residuals displayed a random distribution, indicating the model’s robustness. Consequently, the BATS model was selected for forecasting onion prices spanning from January 2024 to December 2024.