Teardowns and Hedonic Land Value Function Estimation using Non-Sample Information

Teardowns provide direct information on land values in fully developed urban areas because such properties are valued only for their land and location rather than for the characteristics of the structure. Stein-like procedures make efficient use of limited data when a group of variables—the structural characteristics—are expected beforehand to provide little explanatory power. Using data from Chicago for 1995-2003, this study shows how Stein-like rules based on both standard OLS regressions and two-stage selection models can be used to improve land value estimation.