Putting MARS into space. Non linearities and spatial effects in hedonic models

Multivariate Adaptive Regression Spline (MARS) is a simple and powerful non-parametric machine learning algorithm that automatizes the selection of non-linear terms in regression models. In this study, we propose using MARS spatial framework to account for potential non-linearities effects Using relatively large data set 17,000 dwellings St. Petersburg (Russia), examine how works. The empirical evidence shows most explanatory variables model — including lag dependent variable have impact on asking prices dwellings.