We estimate nonmarket values for natural views in an urban setting. These views contain the aesthetics of natural areas commonly found in public parks and open space, and offer an aspect of property valuation that previous research is unable to disentangle from proximity to parks and open space. We incorporate machine learning techniques on Google Street View images to identify natural views in an urban setting. We find positive capitalization rates associated with household views of park-like properties. Estimates are robust to a variety of specifications, including models that are identified off of new developments on neighboring properties and falsification tests that help to rule out the effect of a broader neighborhood environment. From a policy perspective, our results inform as to the optimal size, location, and shape of open space. Furthermore, machine learning methods used in the construction of our view variable provide a potentially powerful tool for other nonmarket valuation studies.

中文翻译：

---

自然景观的舒适价值

我们估算城市环境中自然景观的非市场价值。这些视图包含了公园和开放空间中常见的自然区域的美学，并提供了以前的研究无法将其与公园和开放空间的邻近性分开的房地产估值的一个方面。我们将机器学习技术融入谷歌街景图像中，以识别城市环境中的自然景观。我们发现正资本化率与家庭对公园式房产的看法相关。估计对于各种规格都是稳健的，包括根据邻近房产的新开发项目确定的模型以及有助于排除更广泛的邻里环境影响的伪造测试。从政策角度来看，我们的结果揭示了开放空间的最佳大小、位置和形状。