This study uses yearly data from 1961 to 2018 to forecast milk production in South Asian countries (including China) using ARIMA/GARCH models and Holt’s Linear approach. It is revealed that not all the methods are equally effective in forecasting. Comparison of mean absolute percentage errors between ARIMA and Holt’s Linear model shows that Holt’s approach reveals higher errors.ARIMA forecasting results show that India will be the country with the highest milk production, followed by Pakistan and China while GARCH model fits better to Bangladesh. This paper has policy implications as it can be used for the proper planning of dairy products in the South-Asian counties to safeguard nutritional security.