Out today from SPPD’s Chris Redfearn:

McMillen, D P, and C L Redfearn. “Estimation and Hypothesis Testing for Nonparametric Hedonic House Price Functions.” Journal of Regional Science doi:10.1111/j.1467-9787.2010.00664.x.

In this very nice manuscript, McMillen and Redfearn demonstrate the usefulness of nonparametric estimators on predicting house prices. The method can capture spatial effects such as done here, of local amenities like access to transit. Based on distance from train station in Chicago. The nonparametric estimators allow for a more precise and efficient analysis prediction of housing value based on distance from train stops. On average, they find a 17 percent decline in prices per mile removed from stop, though it’s a bit misleading for me to just throw that finding out since the point of the estimator is to give a more fine-grained indicator of the relationship between housing prices and distances within local areas rather than an average indicator.

Chris Redfearn is one of my absolute favorite colleagues. He does so much interesting work, for one thing, and for another he’s is currently running SPPD’s top-ranked Master’s of Real Estate Development brilliantly.

Add to: Facebook | Digg | Del.icio.us | Stumbleupon | Reddit | Blinklist | Twitter | Technorati | Yahoo Buzz | Newsvine