This study explores the possibilities of applying Google Trends to exchange rate forecasting. Specifically, we construct a sentiment index by using Google Trends to capture market sentiment in Japan and the United States. We forecast the USD/JPY rates using three structural models and two autoregressive models and examine whether our sentiment index can improve the predictive power of these models. We also check the robustness of the main results using the Taylor rule-based model and rolling regression methodology. The data we use run from January 2004 to August 2018, treating January 2004 to February 2011 as the training sample and March 2011 to August 2018 as the forecast sample. We find that the addition of the sentiment index into these models decreases the mean squared prediction error. We also test the sentiment indices of different word numbers and find that the 25- and 30-word indices perform best; in particular, the 30-word index improves all the models tested in this study.

中文翻译：

---

基于谷歌趋势的情绪指数在汇率预测中的应用

本研究探讨了将 Google Trends 应用于汇率预测的可能性。具体来说，我们通过使用谷歌趋势构建情绪指数来捕捉日本和美国的市场情绪。我们使用三个结构模型和两个自回归模型预测美元/日元汇率，并检查我们的情绪指数是否可以提高这些模型的预测能力。我们还使用基于泰勒规则的模型和滚动回归方法检查主要结果的稳健性。我们使用的数据运行时间为2004年1月至2018年8月，将2004年1月至2011年2月作为训练样本，2011年3月至2018年8月作为预测样本。我们发现将情绪指数添加到这些模型中会降低均方预测误差。我们还测试了不同字数的情感指数，发现 25 字和 30 字的指数表现最好；特别是，30 字索引改进了本研究中测试的所有模型。