This paper develops a method to identify three-dimensional anisotropic spatiotemporal autocorrelation with an application to real estate markets. A large literature modeling spatiotemporal autocorrelation in house prices assumes that the spatiotemporal dependence structure is isotropic: a function of only distances between observations, and therefore the direction effect is dismissed. If the importance of direction is dismissed or understated, an estimation result would be biased and therefore less precise unless the distribution of observations is in rare case of being directional homogeneous. This paper thus proposes a local anisotropic spatiotemporal approach to improve estimation performance. The methodology is illustrated by using data on single-family house transactions in the San Francisco Bay Area. The empirical results suggest that the proposed three-dimensional anisotropic modeling technique can reduce both in-sample estimation and out-of-sample forecast errors.
The Journal of Real Estate Finance and Economics – Springer Journals
Published: Aug 14, 2014
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
All the latest content is available, no embargo periods.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud