应用季节周期回归模型预测临床血液需求量

Seasonal regression model applying for forecasting clinical blood demand

目的:应用季节周期回归模型对临床血液需求量进行预测。方法:利用季节周期回归模型对2005～2007年的数据建模,预测2008年各季节临床血液需求量,并将2008年预测值与实际值比较,检验模型的预测能力。结果:对此数据建立了Yij=403225.217-195958.113i+744.767j-183.708sin(ωj)-766.225cos(ωj)模型,平均预测相对误差为10.5%。结论:季节周期回归模型能较好的分析临床血液需求量同季节的关系,并有较强的预测能力,从而为血液中心的库存量预测提供了有效的工具。

Objective： Applying the seasonal regression model for forecasting the clinical blood demand. Methods： Seasonal regression model was applied to establish a model based on the data obtained from the year of 2005 to 2007, and forecasted the seasonal clinical blood demand in 2008, then comparing the predictive value with the actual value in 2008 to test the ability of model. Results： Yij=403225. 217-195958. 113i＋744. 767j-183. 708sin（wj）-766. 225cos（wj） was established for the time series need analysis with an average relative error of 10.5%. Conclusion： Seasonal regression model can analyze the relationship between clinical blood demand and different seasons with better predictive effect , Its strong predictable ability can afford an effective tool for Inventories of blood center.