Forecasting cotton (Gossypium spp.) prices in major Haryana markets: A time series and ARIMA approach

Abstract

Economic outputs are an attractive prospect in any field and hence agriculture also relies heavily on economic stability. The costs associated with cotton farming are increasing and profitability is taking a hit in cotton cultivation. Timely and accurate forecast of the price helps the farmers switch between the alternative nearby markets to sale their produce and getting good prices. Present study was carried out during 2022 to 2023 in Haryana to provide some insights into the possible future prices of cotton (Gossypium spp.) with the help of data collected from AGMARKNET and various major cotton markets (Adampur, Sirsa and Fatehabad) of Haryana. The Autoregressive Integrated Moving Average (ARIMA) models have been employed in order to forecast the prices of cotton crops for the years 2022–23 to 2027–28. Through a meticulous exploration of various combinations of lagged moving average and autoregressive components, the ARIMA (1,1,1) model was selected as the most suitable for the price forecasting in these districts. The results of this analysis demonstrate that the coefficient of determination (R2) for the forecasted cotton crop prices in comparison to the real-time prices falls within acceptable ranges. This finding underscores the efficacy of the ARIMA (1,1,1) model as a reliable tool for generating short-term price estimates. This model offers valuable insights and predictive accuracy, aiding decision-makers and stakeholders in the cotton industry of Adampur, Sirsa and Fatehabad markets to make informed choices and plan effectively for the coming years. Cotton prices vary according to the season and the region, hence a valuable insight on future price assumptions will help the agriculture community.

Keywords

• Autocorrelation, Coefficient of determination, Differencing, Partial autocorrelation function, Price forecast