论文部分内容阅读
目的了解医院住院量的变动趋势,对医院出院人数进行预测分析,为科学决策提供依据。方法应用乘积季节ARIMA模型对某院2003年1月-2013年12月出院人数进行模型拟合,预测2014年各月出院人数,用2014年1月-6月份实际资料评估模型的预测效果。结果该院出院人数呈明显的季节效应,且出院人数逐年小幅递增;乘积季节ARIMA(1,1,1)×(0,1,1)12(不含常数项)模型为最优模型,标准化的BIC(标准化贝叶斯信息量)和平均绝对误差百分比(MAPE)值最小,BIC值为11.98,MAPE值为5.43。Ljung-Box检验无统计学意义(Q18=10.575,P=0.782)。结论乘积季节ARIMA模型可以较好地拟合出院人数的变化趋势,是一种短期预测精度较高的预测模型。
Objective To understand the changing trend of hospitalization and to predict the number of hospital discharges and provide the basis for scientific decision-making. Methods The product ARIMA model was used to model the number of discharged patients from January 2003 to December 2013 in a hospital. The number of discharged patients in each month in 2014 was estimated. The actual data from January to June in 2014 were used to evaluate the predictive effect of the model. Results The number of discharged patients in this hospital showed a significant seasonal effect and the number of discharged patients increased slightly year by year. The product season ARIMA (1,1,1) × (0,1,1) 12 (excluding the constant term) The BIC (standardized Bayesian information) and mean absolute error percentage (MAPE) values were the smallest with a BIC value of 11.98 and a MAPE value of 5.43. Ljung-Box test was not statistically significant (Q18 = 10.575, P = 0.782). Conclusions The product season ARIMA model can better fit the trend of the number of discharged patients and is a prediction model with high short-term prediction accuracy.