基于BP神经网络和ARIMA组合模型的中国入境游客量预测

Forecast of Inbound Tourists to China Based on BP Neural Network and ARIMA Combined Model

游客量的预测和分析是旅游规划与管理的基础性、关键性工作。目前,游客量预测主要采用基于传统研究方法或人工神经网络技术的单项预测方法。近年来的研究表明,组合预测方法比单项预测具有更高的预测精度。本文提出了一种基于BP神经网络和ARIMA组合模型的游客量预测新方法,对中国入境旅游人次数的变化趋势进行了综合分析与预测,预测结果表明这种方法相对于单一的预测方法具有更高的精度,该模型在旅游预测中的应用是可行、有效的。

Forecast and analysis of tourist volume are the basis and key work of tourism planning and management. At present, forecast of tourist volume is mainly based on traditional research approach or single artificial neural network technology. Recent study results show that combined forecast model approach enjoys more precise forecast than monomial forecast approach. The paper proposes a new forecast approach based on BP neural network and ARIMA combined model and makes comprehensive analysis and forecast of the changing trend of inbound tourists to China. Forecast results indicate that this approach is more precise in terms of monomial forecast method. The combined model is feasible and effective in the forecast of overseas tourists.