Forecasting India’s Milk Production using Time-series Econometrics and Machine Learning Models: A Comparative Assessment
Year: 2026
Type: article
Abstract: Background: Milk production is a major component of India’s agricultural economy, contributing significantly to food security, rural livelihoods and economic development. Reliable forecasts of future milk production are essential for effective policy formulation, strategic planning and sustainable growth of the dairy sector. This study evaluated the forecasting performance of time series econometric and machine learning models for predicting India’s milk production. Methods: Annual milk production data for the period 1961-2024 were obtained from the Food and Agriculture Organization Corporate Statistical Database. Time series properties were examined and forecasting models were developed using the Autoregressive Integrated Moving Average, Exponential Smoothing and Neural Network Autoregression approaches. Model performance was evaluated using information criteria, diagnostic tests, forecast accuracy measures and predictive accuracy comparison tests. Result: Milk production increased steadily during the study period, recording a CAGR of 4.05%. The series became stationary after second differencing and the autoregressive integrated moving average (0,2,1) model provided the highest forecasting accuracy, with a root mean square error of 12,693.94, a mean absolute error of 10,407.38 and a mean absolute percentage error of 4.97%. Forecasts indicate that milk production of India will increase from 259.08 million tonnes in 2025 to 323.34 million tonnes by 2030, demonstrating the suitability of the selected model for long-term planning and policy formulation in the dairy sector.
Language: English
Field: Decision Sciences
Domain: Social Sciences
Open Access status: diamond
OpenAlex ID: W7223171072