摘要
目的建立差分自回归滑动平均模型(ARIMA),预测江苏省某三甲医院的平均住院日,为医院医疗资源的配置提供科学参考。方法以江苏省某三甲医院2013年1月至2021年6月全院平均住院日数据为基线,利用R软件构建ARIMA模型,对2021年7月至2022年5月11个月的全院平均住院日进行预测,并与实际值进行对比,评价ARIMA模型的预测效果。结果该医院的平均住院日自2013年1月起逐年呈现下降趋势,并且具有一定的季节特征。拟合最佳的ARIMA模型为ARIMA(0,1,1)(0,1,1)_(12),其平均相对误差MAPE为1.78%,均方根误差RMSE为0.24。ARIMA模型的预测中,RMSE为1.49,MAPE为7.78%,预测结果较为理想。结论应用ARIMA模型对该院的平均住院日预测效果较好,可用于该院平均住院日的短期预测。
Objective To establish a differential autoregressive integrated moving average(ARIMA)model to predict the average length of stay in a tertiary hospital in Jiangsu Province and provide scientific reference for the allocation of medical resources.Methods Based on the average hospital stay data of a tertiary hospital in Jiangsu Province from January 2013 to June 2021,the ARIMA model was constructed by using R software to predict the average hospital stay in the hospital from July 2021 to May 2022,and to compare with the actual value to evaluate the prediction effect of the ARIMA model.Results The average length of stay of the hospital has been decreasing year by year since January 2013,and has certain seasonal characteristics.The best fitted ARIMA model is ARIMA(0,1,1)(0,1,1)_(12),with MAPE of 1.78%and RMSE of 0.24.In the prediction of ARIMA model,RMSE is 1.49,MAPE is 7.78%,and the prediction results are ideal.Conclusion ARIMA model has good prediction effect on the average length of stay in the hospital,and can be used for the short-term prediction of the average length of stay in the hospital.
作者
郭在金
周罗晶
Guo Zaijin;Zhou Luojing(School of Public Health,Yangzhou University,Yangzhou 225009,China;Management Institute,North Jiangsu People′s Hospital,Yangzhou 225001,China)
出处
《中国医院统计》
2022年第4期279-282,共4页
Chinese Journal of Hospital Statistics
关键词
平均住院日
差分自回归滑动平均模型
R语言
预测模型
average length of stay
autoregressive integrated moving average model
R language
prediction model