摘要:
目的:分析宜昌地区ABO血型系统各血型红细胞用量的分布规律,依据时间序列分析方法建立自回归积分滑动平均模型（ARIMA）进行预测,指导采供血机构相关业务工作。方法:在SPSS18.0中利用时间序列模型中专家建模器,对宜昌市2008-01-2015-12红细胞总的用量及各血型分别用量建立数学模型,并预测2016年1至6月用量,与实际用量对比,验证模型误差。结果:专家建模器对红细胞总量、A型及O型红细胞用量给出的模型是ARIMA（0,1,1）（0,1,1）,B型和AB型红细胞用量给出的模型分别是ARIMA（1,1,1）（1,1,1）和ARIMA（2,1,1）（1,1,1）。对5个模型残差的白噪声检验结果均显示P>0.05,说明残差均为白噪声序列,模型提取了原序列中所有数据信息,模型诊断得以通过。将预测值与实际值进行比较,实际值均落入预测值95%的可信区间内,且平均误差相对较小,模型预测效果良好。结论:ARIMA模型能够科学、有效地反映时间序列的变化规律,可以有效预测短期红细胞用量,有针对性地指导血站的采供血业务工作。
Abstract:
Objective:To analyze the distribution of the red blood cell use volume of ABO blood group in Yichang area,and establish the ARIMA model based on time series analysis method to predict the volume and guide the relevant works of the blood collection and supply organizations.Method:We used time series expert modeler in SPSS18.0to set up mathematical models for the monthly total use volume and each kind use volume of red blood cells from January 2008 to December 2015 in Yichang city,and used them to predict the monthly use volume from January to June 2016 and compared them with the actual volume to verify the model error.Result:The expert modeler gave ARIMA（0,1,1）（0,1,1）for the total use volume and A and O type red blood cell use volume,ARIMA（1,1,1）（1,1,1）for B type red blood cell use volume,and ARIMA（2,1,1）（1,1,1）for AB type red blood cell use volume.The residual white noise test results of the five models were all P>0.05,indicating these residuals were white noise and the models extract all the data from the original sequence and passed the test.Comparing the predicted results with the actual values,all of the actual values fell into the predicted interval of 95% confidence level,and their relative errors were small.The predictive effect of the models was perfect.Conclusion:The ARIMA model can scientifically and effectively reflect the changes of time series,effectively predict the short-term use volume of red blood cells,and pertinently guide the blood collection and supply work of blood bank.
下载: