One hundred fifty years of land values in Chicago: A nonparametric approach

Locally weighted regression estimates are presented for Chicago using data from 1836 to 1990. The results demonstrate that nonparametric estimators are useful for hypothesis testing as well as prediction. Estimated regression surfaces for 1836-1928 demonstrate that the monocentric city model continues to adequately represent land values through the early 19th century, but only if the functional form is highly nonlinear. Estimates for 1960-1990 indicate that Chicago is no longer a monocentric city; O'Hare Airport is an important employment center and the blighted areas of the south and west sides have lower land values than the model predicts.