摘要
India is witnessing tremendous growth in dairy industry. The milk production has increased from 20 million tonnes in 1961 to 132 million tonnes in 2012-13. India has been retaining its number one position in milk production for many years. Dairy Industry in India is growing at the rate of 10% per annum. Considering this, it is essential to know the future production to improve and sustain the growth and development of sector. The objective of the study is to find out most suitable forecasting method for milk production for sustainable future production and policy implications. The data used in study is secondary data, collected from FAOSTAT (1961 to 2012) and NDDB (1991 to 2012). Stationarity of data was checked with Autocorrelation Function (ACF) and Partial autocorrelation function (PACF), after confirming the stationarity, Autoregressive Integrated Moving Average (ARIMA) and Vector Autoregression (VAR) models were used. Akaike Information Criteria (AIC), Schwartz Bayesian Criteria (SBC), Mean Absolute Percentage Error (MAPE), R square and RMSE were used to test reliability of model. The results indicate that ARIMA (1, 1, 1) is more suitable method with the use of SPSS software package for forecasting of milk. Milk production is expected to be 160 million tonnes by 2017.
摘要译文
印度目睹了乳制品行业的巨大增长。牛奶产量从1961年的2000万吨增加到2012 - 13年的1.32亿吨。印度一直保留了许多年份牛奶产量的头号职位。印度的乳制品行业以每年10％的速度增长。考虑到这一点，必须了解未来的生产，以改善和维持部门的增长和发展。该研究的目的是找出可持续未来生产和政策影响的最适合牛奶产量的预测方法。研究中使用的数据是来自Faostat（1961年至2012）和NDDB（1991年至2012年）收集的二级数据。通过自相关函数（ACF）和部分自相关函数（PACF）检查数据的实用性，并在确认实用性，自动增加的综合移动平均（ARIMA）和向量自动增加（VAR）模型之后。 Akaike信息标准（AIC），Schwartz贝叶斯标准（SBC），平均绝对百分比误差（MAPE），R Square和RMSE用于测试模型的可靠性。结果表明，使用SPSS软件包以预测牛奶，ARIMA（1,1,1）更合适的方法。到2017年，牛奶产量预计将达到1.6亿吨。
Sagar Surendra Deshmukh*[1];R. Paramasivam. Forecasting of milk production in India with ARIMA andVAR time series models[J]. Asian Journal Of Dairy and Food Research, 2016,35(1): 17-22