HEDONIC ANALYSIS OVER TIME AND SPACE: THE CASE OF HOUSE PRICES AND TRAFFIC NOISE

One consequence of the expanding road network and its associated traffic is increased levels of traffic noise. While the hedonic literature has consistently found a negative relationship between real estate prices and noise levels, research in the United States has typically relied on crude measures of traffic noise. Here, we reduce the measurement error of traffic noise exposure through a detailed model of noise propagation over the landscape. We then estimate the hedonic relationship between noise and single family house prices using over 40,000 transactions throughout the St. Paul, Minnesota, urban area from 2005 to 2010. We implement spatially and temporally flexible local regression techniques and find significant nonstationarity in the hedonic function over time and space.