季节性组合预测模型在医院门诊量中的应用研究

医院门诊量是一个具有复杂的非线性组合特征的季节性时问序列。为解决传统时间序列预测大多数都是使用单一模型，以致影响预测精度等问题，采用了最优加权组合预测方法将季节性ARIMA乘积模型和季节性神经网络模型进行组合优化。结果表示，季节性组合预测模型在拟合精度和预测准确性方面优于任何单一预测方法，为季节性时间序列预测提供了一种新的实用方法。

Application research of seasonal combined forecasting model in hospital outpatient amount

The hospital outpatient amount is a seasonal time series with the character of complex non-linear combination. A seasonal combined forecasting model based on multiple seasonal ARIMA model and seasonal ANN model was put forward to overcome the defaults when traditional forcasting of single model was mostly used so as to affect the prediction precision. These single models were applied to build the optimal mix forecasting method. Prediction result shows the model is superior to the singular one in many performance aspects. It offers a new effective solution of seasonal time series prediction.