摘要
目的对某院2011年1月-2016年5月乳腺外科出院人数进行时间序列预测分析,旨在为医院信息化、科学化管理提供参考依据,同时对ARIMA模型和Holt-Winters模型进行分析比较。方法通过ARIMA模型和Holt-Winters模型对乳腺外科出院人数进行模型训练,然后利用2016年6月、7月出院人数进行验证,并利用所建立的模型对2016年8月-12月出院人数进行预测。通过ARIMA模型和Holt-Winters模型对出院人数数据进行拟合,并与2016年6月和7月实际值进行比较。结果 ARIMA(0,1,2)(1,1,0)_(12)模型中非季节性移动平均系数MA1=-0.9549,季节性自回归系数SAR1=-0.3278。Box-Ljung统计量检验均无统计学意义(P=0.148),因此残差序列为白噪声序列。Holt-Winters模型对时间序列数据进行对数变换修正后的α,β,γ的估计值分别为0.19,0.31,0.76。Box-Ljung检验统计量无统计学意义(P=0.88),意味着延迟1~20阶为非零自相关,并且预测误差在整个时间段呈现出服从零均值,方差不变的正态分布。结论乳腺外科出院人数存在逐年上升的趋势,并且具有很明显的季节性趋势,Holt-Winters模型相比ARMIA模型对于短期的出院人数预测具有更强的实用价值,可作为科学管理提供参考依据。
Objectives To conduct time series analysis on the discharged patients number in breast cancer department from January 2011 to May 2016 of a hospital,so as to provide a reference for the informatization and scientific management of a hospital.Methods ARIMA model and Holt winters model were adopted to January 2011 to2016 may breast surgery discharge number of patients to train the model,then discharged the number was verified by2016 in June and July,and the number of discharged patients from August to December 2016 was predicted by the established model.The discharge data were fitted by ARIMA model and Holt-Winters model,and the data were compared with the actual value in July and June 2016.Results ARIMA（0,1,2）（l,1,0） 12 model of non-seasonal moving average coefficient MA1=-0.9549,seasonal auto regressive coefficient SAR1=-0.3278.Box-Ljung statistic test was not statistically significant（P=0.148）,so the residual sequence was white noise sequence.Holt-Winters model for the time series data of the modified alpha,beta,gamma estimates were 0.19,0.31,0.76,respectively.Box-Ljung test statistic was not statistically significant（P=0.88）,which meant that the delay 1-20 order was.non zero autocorrelation,and the prediction error in the whole time period was presented with the zero mean,the variance of the normal distribution.Conclusions The number of discharged patients had a trend toward increasing year by year in breast surgery department,and had obvious seasonal trend.The Holt winters model had more practical value compared with the model of ARMIA for short-term discharge number prediction,so it could provide a reference for scientific management.
作者
杨博文
迟卫军
付凌雨
苏芃
Yang Bowen Chi Weijun Fu Lingyu Su Fan
出处
《中国病案》
2017年第1期50-54,共5页
Chinese Medical Record
