基于修正动态组合模型对公共交通客运量的预测

Forecast of public transport passenger volume based on modified dynamic combination model

针对公共交通客运量的预测问题,结合ARIMA、灰色预测以及BP神经网络的优势,采用临近期的误差平方和来计算动态权重,将突发事件定义为影响因子,建立了修正动态加权组合模型.选取北京市1978—2021年公共交通客运量进行实证分析.实证分析结果表明,修正动态加权组合模型的预测效果比单一模型和固定权重组合模型更好.

Aiming at the forecasting problem of public transport passenger volume,combining the advantages of ARIMA,grey prediction and BP neural network,the dynamic weight is calculated by the sum of the squares of errors in the near period,the modified dynamic weighted combination model is established by defining emergency events as impact factors.The public transport passenger volume of Beijing from 1978 to 2021 is selected for empirical analysis.The empirical analysis results show that the modified dynamic weighted combination model has better prediction effect than the single model and the fixed weight combination model.