Forecasting Cotton Productivity Of Haryana And Punjab Bysuitable Arima Model

ABSTRACT The primary goal of this research is to determine which Auto-Regressive Integrated Moving Average (ARIMA) model is best for predicting cotton productivity in Punjab and Haryana. Time series data spanning the last ten years, from 2012–13 to 2021–22, has been studied for this study. The Normalized Bayesian Information Criterion (BIC), Mean Absolute Percentage Error (MAPE), and maximum R-squared (R²) values were used to calculate which models were the best. The findings indicate that ARIMA (1,1,1) is the best-fit model for Haryana, while ARIMA (2,1,1) is the most suitable for Punjab. The analysis suggests a declining trend in cotton productivity for both Haryana and Punjab