Spatial and hedonic analysis of house price dynamics in Warsaw

The aim of our article is to analyze the dynamics of housing prices in the secondary housing market in Warsaw from Q1 2006 to Q3 2013, taking into account the spatial relationship between prices. In the first part of this research we compare the geographically weighted regression with a linear regression estimated using OLS with spatial variables. In the second part, we combine the geographically weighted regression with the penalized spline regression to extract the effect of time on prices. With this method we obtain a nonlinear and more precise measure of time effects and improved goodness-of-fit statistics. We obtain a hedonic index, that is more robust against short-term changes in house prices than the usual, linear hedonic index. This is a novel approach, which has not been applied before in the case of the Polish housing market. The index allows us to show how interest rates or the housing policy influenced house prices.