Vol. 14, Issue 4 (2025)

Author(s):
Krishnaveni G, Kuldeep Tandan, ML Lakhera and Sweta Ramole

Abstract:
Accurate forecasting of grain yield and straw yield of rice crop is essential for ensuring food security and sustainable agricultural planning. This study investigates the application of the Autoregressive Integrated Moving Average (ARIMA) model for predicting grain yield and straw yield of rice crop in long-term fertilizer experiments (LTFEs). The research utilizes historical yield data from LTFEs, incorporating different fertilizer treatments to analyze trends and patterns. The ARIMA model is employed to assess time-series data, identify optimal model parameters, and generate forecasts. Model performance is evaluated using statistical measures such as Mean Absolute Error (MAE) and Root Mean Square Error (RMSE). The analysis for rice grain yield and rice straw yield identified ARIMA (0,1,0) and ARIMA (2,2,0) as the optimal model specification respectively. By applying this model to the time series data, projections were made for the next five years with the objective of achieving the highest possible forecasting accuracy.