论文部分内容阅读
为了通过对牛群的合理调整达到去产能、去库存的目标,试验基于2008年7月份—2015年12月份的967 767头次奶牛月平均产奶量数据,建立了自回归积分滑动平均模型,以预测未来24个月宁夏地区奶牛平均产奶量。结果表明:对模型进行检验后得出AIC值为210.83,Ljung-Box测试显示模型残差P>0.05,说明残差为白噪声,可对未来24个月的每月平均产奶量进行预测;未来24个月奶牛平均产奶量总体依然呈现上升趋势,但也有波动;每年5,6,7月份为产奶量最高月份,1,2,3月份为产奶量最低月份。
In order to meet the target of de-capacity and stock removal through the rational adjustment of herd, the test was based on the average monthly milk production of 967 767 dairy cows from July 2008 to December 2015, and established the autoregressive integral sliding average model. To predict the average milk production in Ningxia in the next 24 months. The results showed that the AIC value of the model was 210.83. The Ljung-Box test showed that the model residual P> 0.05, indicating that the residual noise was white noise, which could predict the average monthly milk production in the next 24 months. In the next 24 months, the average milk yield of dairy cows still shows an upward trend but fluctuates; the highest milk yield is obtained in May, June and July each year, and the lowest milk yield in January, February and March.