House prices vary with location. At the same time the border between two neighboring housing markets tends to be fuzzy. When we seek to explain or predict house prices we need to correct for spatial price variation.
A much used way is to include neighborhood dummy variables. In general, it is not clear how to choose a spatial subdivision in the vast space of all possible spatial aggregations. We take a biologically inspired approach, where diﬀerent spatial aggregations mutate and recombine according to their explanatory power in a standard hedonic housing market model. We ﬁnd that the genetic algorithm consistently ﬁnds aggregations that outperform conventional aggregation both in and out of sample.
A comparison of best aggregations of diﬀerent runs of the genetic algorithm shows that even though they converge to a similar high explanatory power, they tend to be genetically and economically diﬀerent. Diﬀerences tend to be largely conﬁned to areas with few housing market transactions.
The paper is published as a CLTS Working Paper and can be downloaded here.
CLTS WP 04/18
Published 24. April 2018 - 12:12 - Updated 24. April 2018 - 12:16