Data-Driven Estimation of Treatment Buffers in Hedonic Analysis: An Examination of Surface Coal Mines

Abstract

In hedonic studies of how environmental disamenities influence house prices, the homes that are treated are typically unknown. We propose a new method for defining treatment buffers and apply our technique to disamenities that have thus far received little attention: surface coal mines. Our leave-one-out cross-validation approach identifies an optimal buffer of 2,300 m in two Appalachian counties at which effects dissipate. Hedonic regressions indicate that treated homes sell for 15.5% less than untreated homes. We supplement these results with nearest-neighbor covariate matching models and recover average treatment effects on the treated of 14.7% price reductions.