Selection of the Best Time Series Arima Model to Forecast Onion Production in Bangladesh

Arima Forecasting of Onion Production in Bangladesh

Abstract

Onion is one of the most commonly used spices in Bangladesh for preparing food and illness treatment. Since the demand for onions is rising daily, accurate projections must be made to implement policies based on that need. Hence, the study attempts to select the best Auto-Regressive Integrated Moving Average (ARIMA) model to forecast onion production in Bangladesh. Initially, the study considers Bangladesh's yearly onion production (in hg/ha) data set from 1971 to 2017. The study then applied the Box-Jenkins approach to build the model and forecast the onion production in Bangladesh. The study also used data from 2018 to 2022 to compare the forecasted values. The AIC, AIC_C, and BIC values, error metrics, and residuals of the fitted model were assessed using onion production data. The comparative analysis shows that ARIMA (0, 2, 1) is the best model for precise forecasts of onion production in Bangladesh. The proposed model may help policymakers forecast accurately and know the future trend of onion production in Bangladesh.

SAARC J. Agric., 22(2): 169-180 (2024)