旅游作为绿色经济推动了地区经济社会的发展。本文以象鼻山景区为例,利用百度指数分析游客对该景点的网络关注度,并针对单一模型对景区日客流量预测精度不足的问题展开研究。提出将ARIMAX模型与GA-XGBoost模型采用残差法进行组合,将数...旅游作为绿色经济推动了地区经济社会的发展。本文以象鼻山景区为例,利用百度指数分析游客对该景点的网络关注度,并针对单一模型对景区日客流量预测精度不足的问题展开研究。提出将ARIMAX模型与GA-XGBoost模型采用残差法进行组合,将数理统计模型和机器学习模型组合,实现优势互补,提高预测精度;首先使用ARIMAX对数据进行预测分析,称预测结果为y1,再把ARIMAX模型的残差放入XBGoost模型进行学习,基于GA算法对XGBoost的超参数进行优化,解决了ARIMAX模型难以对非线性数据预测的问题,GA-XGBoost的预测结果为y2,组合模型的最终预测结果y=y1+y2。最后,根据预测误差评价指标对多个模型进行对比。研究结果表明,ARIMAX-GA-XGBoost组合模型预测精度更高,适应性及泛化能力更强,可为旅游相关管理部门的科学决策提供必要的参考,具有很高的经济效益与实际意义。Tourism, as a green economy, drives regional socioeconomic development. Taking Xiangbi Mountain Scenic Area as a case study, this paper analyzes the network attention of tourists towards this attraction using Baidu Index. To address the issue of insufficient prediction accuracy of single models for daily tourist flow forecasting in scenic areas, a hybrid modeling approach is proposed. By integrating the ARIMAX model with the GA-XGBoost model through residual combination methodology, this study combines mathematical-statistical modeling with machine learning techniques to achieve complementary advantages and enhance prediction accuracy. Specifically, the ARIMAX model is initially employed for data prediction (denoted as y1), followed by feeding its residuals into the XGBoost model for learning. The genetic algorithm (GA) optimizes XGBoost’s hyperparameters, effectively resolving ARIMAX’s limitations in handling nonlinear data prediction (GA-XGBoost prediction denoted as y2). The final combined model output is formulated as y=y1+y2. Comparative analysis of multiple models through prediction error evaluation metrics demonstrates that the ARIMAX-GA-XGBoost hybrid model exhibits superior prediction accuracy, enhanced adaptability, and stronger generalization capabilities. This research provides valuable decision-making references for tourism management authorities and holds significant economic benefits and practical implications.展开更多
ARIMAX模型在多个领域有着重要应用。但是针对ARIMAX模型的参数估计均是经典统计方法,利用贝叶斯估计是一个值得探究的问题。考虑到共轭先验分布的性质特点,对于ARIMAX模型中的系数的先验分布为正态分布,噪声项的方差先验分布为逆伽马...ARIMAX模型在多个领域有着重要应用。但是针对ARIMAX模型的参数估计均是经典统计方法,利用贝叶斯估计是一个值得探究的问题。考虑到共轭先验分布的性质特点,对于ARIMAX模型中的系数的先验分布为正态分布,噪声项的方差先验分布为逆伽马分布假定,本文给出了参数的后验分布,并使用Gibbs采样的方式,给出各个参数的一个估计。模拟试验的结果表明,本文的估计方法具有很好的功效,借助参数的迭代图表明文中使用的方法具有稳健性。使用创业板数据与上证数据做出实证分析,并发现文中给出的方法不仅具有很好的解释性,同时能够提取出完整的数据信息。The ARIMAX model has important applications in various fields. However, parameter estimation for the ARIMAX model has traditionally relied on classical statistical methods, and exploring Bayesian estimation presents a worthwhile direction. Considering the properties of conjugate prior distributions, this study assumes a normal prior distribution for the coefficients in the ARIMAX model and an inverse gamma prior for the variance of the noise term. The posterior distributions of the parameters are derived, and estimates for each parameter are obtained using Gibbs sampling. Simulation results demonstrate the effectiveness of the proposed estimation method. Iterative plots of the parameters indicate the robustness of the method. An empirical analysis using data from the ChiNext Index and the Shanghai Composite Index further reveals that the proposed method not only offers strong interpretability but also effectively captures comprehensive information from the data.展开更多
目的探讨比较多元自回归移动平均模型(Autoregressive Integrated Moving Averagemodel-X,ARIMAX)与多变量长短期记忆神经网络(Long Short Term Memory Network,LSTM)模型在盐城市总死亡人数预测中的效果。方法采用2014年1月1日至2017年...目的探讨比较多元自回归移动平均模型(Autoregressive Integrated Moving Averagemodel-X,ARIMAX)与多变量长短期记忆神经网络(Long Short Term Memory Network,LSTM)模型在盐城市总死亡人数预测中的效果。方法采用2014年1月1日至2017年6月30日江苏省盐城市每日总死亡人数、气象因素和空气质量数据,建立ARIMAX及多变量LSTM模型,并对2017年7月1日至7月14日每日总死亡人数进行预测,以RMSE、MAE、MAPE为评价指标比较两种模型的预测效果。结果ARIMAX(4,1,1)模型和多变量LSTM模型的RMSE、MAE、MAPE值分别为20.742、15.094、9.921和47.182、35.863、19.633。结论ARIMAX模型比多变量LSTM模型更适于预测盐城市每日死亡人数。展开更多
文摘旅游作为绿色经济推动了地区经济社会的发展。本文以象鼻山景区为例,利用百度指数分析游客对该景点的网络关注度,并针对单一模型对景区日客流量预测精度不足的问题展开研究。提出将ARIMAX模型与GA-XGBoost模型采用残差法进行组合,将数理统计模型和机器学习模型组合,实现优势互补,提高预测精度;首先使用ARIMAX对数据进行预测分析,称预测结果为y1,再把ARIMAX模型的残差放入XBGoost模型进行学习,基于GA算法对XGBoost的超参数进行优化,解决了ARIMAX模型难以对非线性数据预测的问题,GA-XGBoost的预测结果为y2,组合模型的最终预测结果y=y1+y2。最后,根据预测误差评价指标对多个模型进行对比。研究结果表明,ARIMAX-GA-XGBoost组合模型预测精度更高,适应性及泛化能力更强,可为旅游相关管理部门的科学决策提供必要的参考,具有很高的经济效益与实际意义。Tourism, as a green economy, drives regional socioeconomic development. Taking Xiangbi Mountain Scenic Area as a case study, this paper analyzes the network attention of tourists towards this attraction using Baidu Index. To address the issue of insufficient prediction accuracy of single models for daily tourist flow forecasting in scenic areas, a hybrid modeling approach is proposed. By integrating the ARIMAX model with the GA-XGBoost model through residual combination methodology, this study combines mathematical-statistical modeling with machine learning techniques to achieve complementary advantages and enhance prediction accuracy. Specifically, the ARIMAX model is initially employed for data prediction (denoted as y1), followed by feeding its residuals into the XGBoost model for learning. The genetic algorithm (GA) optimizes XGBoost’s hyperparameters, effectively resolving ARIMAX’s limitations in handling nonlinear data prediction (GA-XGBoost prediction denoted as y2). The final combined model output is formulated as y=y1+y2. Comparative analysis of multiple models through prediction error evaluation metrics demonstrates that the ARIMAX-GA-XGBoost hybrid model exhibits superior prediction accuracy, enhanced adaptability, and stronger generalization capabilities. This research provides valuable decision-making references for tourism management authorities and holds significant economic benefits and practical implications.
文摘ARIMAX模型在多个领域有着重要应用。但是针对ARIMAX模型的参数估计均是经典统计方法,利用贝叶斯估计是一个值得探究的问题。考虑到共轭先验分布的性质特点,对于ARIMAX模型中的系数的先验分布为正态分布,噪声项的方差先验分布为逆伽马分布假定,本文给出了参数的后验分布,并使用Gibbs采样的方式,给出各个参数的一个估计。模拟试验的结果表明,本文的估计方法具有很好的功效,借助参数的迭代图表明文中使用的方法具有稳健性。使用创业板数据与上证数据做出实证分析,并发现文中给出的方法不仅具有很好的解释性,同时能够提取出完整的数据信息。The ARIMAX model has important applications in various fields. However, parameter estimation for the ARIMAX model has traditionally relied on classical statistical methods, and exploring Bayesian estimation presents a worthwhile direction. Considering the properties of conjugate prior distributions, this study assumes a normal prior distribution for the coefficients in the ARIMAX model and an inverse gamma prior for the variance of the noise term. The posterior distributions of the parameters are derived, and estimates for each parameter are obtained using Gibbs sampling. Simulation results demonstrate the effectiveness of the proposed estimation method. Iterative plots of the parameters indicate the robustness of the method. An empirical analysis using data from the ChiNext Index and the Shanghai Composite Index further reveals that the proposed method not only offers strong interpretability but also effectively captures comprehensive information from the data.
文摘目的探讨比较多元自回归移动平均模型(Autoregressive Integrated Moving Averagemodel-X,ARIMAX)与多变量长短期记忆神经网络(Long Short Term Memory Network,LSTM)模型在盐城市总死亡人数预测中的效果。方法采用2014年1月1日至2017年6月30日江苏省盐城市每日总死亡人数、气象因素和空气质量数据,建立ARIMAX及多变量LSTM模型,并对2017年7月1日至7月14日每日总死亡人数进行预测,以RMSE、MAE、MAPE为评价指标比较两种模型的预测效果。结果ARIMAX(4,1,1)模型和多变量LSTM模型的RMSE、MAE、MAPE值分别为20.742、15.094、9.921和47.182、35.863、19.633。结论ARIMAX模型比多变量LSTM模型更适于预测盐城市每日死亡人数。
