This study draws on machine learning as a means to causal inference for econometric investigation. We utilize the concept of transfer entropy to examine the relationship between the US National Association of Home Builders Index and the S&P CoreLogic Case-Shiller 20 City Composite Home Price Index (SPCS20). The empirical evidence implies that the survey data can help to predict US house prices. This finding extends the results of Granger causality tests performed by Rodriguez Gonzalez et al. in 2018 using a new machine learning approach that methodologically differs from traditional methods in empirical financial research.

中文翻译：

---

美国房价领先指标：对欧盟金融风险管理者的新证据和启示

本研究利用机器学习作为计量经济学调查因果推断的一种手段。我们利用转移熵的概念来检验美国全国房屋建筑商协会指数与标准普尔 CoreLogic Case-Shiller 20 城市综合房价指数 (SPCS20) 之间的关系。经验证据表明，调查数据有助于预测美国房价。这一发现扩展了 Rodriguez Gonzalez 等人进行的格兰杰因果检验的结果。在 2018 年使用了一种新的机器学习方法，该方法在方法论上不同于实证金融研究中的传统方法。