Housing price gradients in a geography with one dominating center

We primarily focus on explaining housing prices and predicting housing price gradients in a Norwegian region with one dominating center (Stavanger). For such a geography spatial separation can be represented in a hedonic regression equation by a function of traveling distance from the city center. Several functions are tested, and some alternatives provide both a satisfying goodness-to-fit, consistent coefficient estimates, and intuitively reasonable predictions of housing price gradients. Still, not all commonly used functions are recommended. Spatial autocorrelation is removed when the hedonic function is properly specified.