1. Rural agro-advisory services are in the process of dynamic transformation across the globe. Interventions of different information and communication technology (ICT) modules in the rural advisor...

The present study was conducted to understand the present and future perspective of milk production in Karnataka and to develop a suitable model for prediction of milk production in the state. The present study is having advantage as the past trends of milk production in the state have been impressive with growth, contributing significantly to overall milk production in the country. The Auto-Regressive Integrated Moving Average (ARIMA) and Artificial Neural Network (ANN) models were applied to the data sets of different dairy units for modeling and forecasting milk production in the state. The milk production of different dairy units has indicated a significant and positive trend over the years (Mann Kendal and Sen’s estimator). The purpose of the study was based on the accuracy criteria to identify the best ARIMA and ANN models for forecasting. It was found that the ARIMA model predicted cows, goats, and total milk production with better R-Square, MAPE and percentage prediction error compared to ANN models. ANN model performed better in predicting the buffalo’s milk yield compared to the ARIMA model. The best-identified models were used for out-sample forecast.