恶性肿瘤住院量与住院费用的ARIMA乘积季节模型预测研究

ARIMA Product Season Model for Predicting Number of Inpatient and Hospitalized Expense of Malignant Tumor

目的探讨自回归求和移动平均(ARIMA)乘积季节模型在恶性肿瘤住院量与住院费用中的应用,为医院恶性肿瘤业务管理提供科学依据。方法收集某院2007-2016年逐月恶性肿瘤住院患者资料,采用ARIMA乘积季节模型对2007-2015年逐月恶性肿瘤的住院人次和住院费用进行模型拟合,用2016年逐月数据评价其预测效果,并预测2017年恶性肿瘤逐月住院人次与住院费用。结果 ARIMA(0,1,1)(1,1,0)_(12)是恶性肿瘤住院人次与住院费用的最佳拟合预测模型,拟合相对误差分别为1.1%和1.47%。根据ARIMA(0,1,1)(1,1,0)_(12)预测结果,2017年恶性肿瘤住院量将达7631人次,住院费用将达3.36亿元。结论 ARIMA季节乘积模型能很好地应用于医院业务管理预测中。

Objective To explore the application of auto-regressive integrated moving average （ARIMA） product sea- son model in predicting number of inpatient and hospitalized expense of malignant tumor, and to provide scientific basis for hos- pital business management. Methods We collected inpatient data of malignant tumor from January 2007 to December 2015 in one hospital for model fitting,and used monthly data 2016 to verify the effect of model prediction. We predicted the number of inpatient and hospitalized expense of malignant tumor in 2017. Results ARIMA （ 0,1,1 ） （ 1,1,0 ） ~2 was the best model for number of inpatient and hospitalized expense of malignant tumor, with prediction fitting errors of 1.1% and 1.47 %, respectively. The number of inpatient and hospitalized expense of malignant tumor in 2017 were predicted to be 7631 and 0. 336 billion. Conclusion ARIMA product season model can better applied in the predicting of hospital business management.