本书旨在探讨气候变量对旅游需求的影响以及引入气候变量是否可以提高旅游需求预测精度。文章系统评述了包括组合预测方法、计量经济模型、单变量时间序列模型、人工智能技术在内的各种旅游需求预测方法及模型，综述发现，已有研究忽视气候变量在旅游需求分析及预测中的作用。在此基础上，作者提出新的旅游需求预测方法：即引入旅游气候指数的组合预测策略。实证研究发现，旅游气候指数对旅游需求存在显著的积极影响，新的预测策略能够提高预测稳定性及精度。

吴曦，经济学博士，研究方向为旅游需求预测与旅游可持续发展。现任中原工学院经济管理学院副教授，中原工学院系统与工业工程技术研究中心研究员。2007年6月毕业于北京师范大学，获经济学硕士学位，同年7月进入中原工学院经济管理学院任教。2015年10月获得Erasmus Mundus全额博士奖学金，赴伯恩茅斯大学（Bournemouth University）攻读博士学位。读博期间，先后受聘于伯恩茅斯大学担任兼职讲师及兼职研究员，2017年获得第六届国际旅游经济学协会 （International Association for Tourism Economics，IATE）国际双年会博士奖学金及最佳论文奖。2020年1月博士毕业，获经济学博士学位。

Chapter 1 Introduction
1.1 Research Background and Motivation
1.2 Research Aim and Objectives
1.3 Overview of International Demand for UK Tourism
1.4 Structure of This Book
Chapter 2 Literature Review
2.1 Introduction
2.2 Non-causal Time Series Techniques
2.2.1 Overview
2.2.2 The Na?ve Model
2.2.3 The Autoregressive Integrated Moving Average Model
2.2.4 The Exponential Smoothing Model
2.2.5 The State Space ETS Model
2.2.6 The Structural Time Series Model
2.2.7 The Singular Spectrum Analysis
2.3 Causal Econometric Models
2.3.1 Overview
2.3.2 The Single - Equation Approach
2.3.3 The System - of - Equations Approach
2.3.4 Studies Based on Panel Data
2.3.5 Climate and Tourism Demand
2.4 The Combination Forecasting Approach
2.4.1 Why to Combine
2.4.2 Weighting Schemes
2.4.3 Applications in the Tourism Demand Literature
2.5 Data
2.5.1 Data Type
2.5.2 Data Frequency
2.6 Summary
Chapter 3 Research Method
3.1 Introduction
3.2 Research Plan
3.3 Variables and Data
3.3.1 The Dependent Variable
3.3.2 Explanatory Variables
3.3.3 Data Sources
3.3.4 Data Sample
3.4 The Bounds Test Cointegration Approach
3.5 Diagnostic Tests
3.5.1 The Jarque – Bera Normality Test
3.5.2 The Breusch – Godfrey Lagrange Multiplier Test
3.5.3 Testing for Heteroscedasticity
3.5.4 The Ramsey Regression Equation Specification Error Test
3.6 Forecasting Methods
3.6.1 Individual Forecasting Models
3.6.2 The Combination Forecasting Approach
3.7 Forecasting Procedures and Accuracy Measures
3.7.1 The Recursive Individual Forecasting Procedure
3.7 .2 The Recursive Weighting Procedure
3.7.3 Forecasting Accuracy Measures
3.8 Programming for Combination Forecasting
Chapter 4 Results and Discussion: Impact Analysis
4.1 Introduction
4.2 Results of Unit Root Tests
4.3 Results of Bounds Test
4.4 Estimation Results and the Impact of Influencing Factors
4.4.1 Income
4.4.2 Own Price
4.4.3 Substitute Price
4.4.4 One-off Events
4.4.5 The Climate Factor
4.4.6 The Error Correction Term
4.4.7 Diagnostic Tests
4.5 Summary
Chapter 5 Results and Discussion: Forecasts Comparison
5.1 Introduction
5.2 Forecasting Performance of Individual Models
5.2.1 Comparison for Different Origins
5.2.2 Comparison for Different Forecasting Horizons
5.2.3 General Comparison among Various Individual Models
5.2.4 Whether Including the Climate Factor Can Improve the Forecasting Ability of Econometric Models
5.3 Forecasting Performance of Combination Methods
5.3.1 Combining All Models
5.3.2 Combining Traditional Econometric and Time Series Models
5.3.3 Combining Climate Econometric and Time Series Models
5.3.4 Comparison among Three Combination Groups
5.4 Which Models to Combine
5.5 How Many Models to Combine
5.6 Conclusion
Chapter 6 Conclusion
6.1 Introduction
6.2 Summary of the Findings
6.3 Limitations of the Current Research
6.4 Recommendations for Future Research
Bibliography