The Baidu Index is one of the major sources of big data on the internet. By leveraging the characteristics of big data, it can enhance the predictive capability of models in forecasting tourist volumes. Using Hangzhou as a case study, this paper selects the Baidu Index as the independent variable and daily tourist numbers in Hangzhou, with a sample size of 1562, as the dependent variable. Two types of independent variables were constructed through index synthesis and lagged correlation analysis to build LSTM and ARDL models for comparison of predictive performance. The results indicate that: (1) There is a long-term relationship and Granger causality between daily tourist volumes in Hangzhou and the Baidu Index. (2) The ARDL (1,5,2,0) model, using high-correlation terms as independent variables, achieves the best predictive performance with an RMSE of 7.95 and a goodness-of-fit of 84.83%. (3) The predicted data can provide a basis for government decision-making, helping to manage peak tourist seasons rationally and promoting the revival of the tourism industry.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Prediction of Daily Tourist Volume in Hangzhou Based on Baidu Index

  • Baoxin Si,
  • Qing Ma,
  • Yunkai Yang,
  • Zheyuan Li

摘要

The Baidu Index is one of the major sources of big data on the internet. By leveraging the characteristics of big data, it can enhance the predictive capability of models in forecasting tourist volumes. Using Hangzhou as a case study, this paper selects the Baidu Index as the independent variable and daily tourist numbers in Hangzhou, with a sample size of 1562, as the dependent variable. Two types of independent variables were constructed through index synthesis and lagged correlation analysis to build LSTM and ARDL models for comparison of predictive performance. The results indicate that: (1) There is a long-term relationship and Granger causality between daily tourist volumes in Hangzhou and the Baidu Index. (2) The ARDL (1,5,2,0) model, using high-correlation terms as independent variables, achieves the best predictive performance with an RMSE of 7.95 and a goodness-of-fit of 84.83%. (3) The predicted data can provide a basis for government decision-making, helping to manage peak tourist seasons rationally and promoting the revival of the tourism industry.