Forecasting buffalo milk production in India: Time series approach

Abstract

Globally, India stands first both in production as well as consumption of milk. Nearly 50% of the Indian milk production comes from buffalo followed by cow and goat. India is home to buffalo with approximately 56% of world buffalo population. Given the importance of buffalo milk, an attempt was made to model and forecast the annual buffalo milk production using various time series analysis techniques. The Autoregressive Integrated Moving Average (ARIMA) model, the Artificial Neural Network (ANN) model and ARIMA-ANN hybrid model were used to model the time series data of 58 years collected from secondary sources. Among the three models, the ARIMA-ANN hybrid model was found to be the best for the data under consideration based on forecast accuracy measures. This is because of the ability of ARIMA-ANN hybrid model to capture both linear and nonlinear structures in the data. By using the ARIMA-ANN hybrid model, the annual buffalo milk production was forecasted and it was found to exceed 1,000 million tonnes in the coming years.