An application of univariate model for forecasting cotton production in Maharashtra

Cotton is one of the most important fibers and cash crops in Maharashtra, and it plays a major part in the state's industrial and agricultural economy. It provides the cotton textile industry with the basic raw material (cotton fiber). In the current study, data on cotton production in Maharashtra were gathered from the Cotton Corporation of India website and used to fit the ARIMA model to forecast future production. The time series data was collected from 1964-65 to 2024-25. For forecasting, the Box Jenkins (1970) ARIMA methodology has been applied. The most well-liked and frequently employed forecasting model for time series data is ARIMA. For the gathered data, autocorrelation and partial autocorrelation functions were constructed. R programming software was used to estimate model parameters. The performance of the fitted model was examined by computing various measures of goodness of fit viz., AIC, AICc, BIC, and MAPE. Empirical results showed that the ARIMA (0,1,3) model was most suitable to forecast the future production of cotton in Maharashtra. The forecasts from 2025-2026 to 2030-2031 are calculated based on the selected model. Overall cotton production is expected to be 93.38 million tons by 2031. The forecasting power of the autoregressive integrated moving average model was used to forecast cotton production for six leading years. The results of cotton production in Maharashtra are presented numerically and graphically.