Cointegration in Non-Stationary Spatial Panel Data

Michael Beenstock*, Daniel Felsenstein

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingChapterpeer-review

1 Scopus citations


This chapter uses Israeli data to illustrate the estimation of spatially cointegrating vectors using the methodology presented in Chap. 7. Since the empirical illustration is for house prices and housing construction in Israel, a theoretical model is proposed in which the housing markets in two regions are related. The model jointly determines the spatiotemporal dynamics for house prices and housing construction in the two regions. We begin with cointegrating vectors for single equations for house prices in which in the long run regional house prices are hypothesized to vary directly with regional income and population, and inversely with their housing stocks, as well as spatial lags of these variables. Results show that house prices are globally cointegrated with these variables. This is followed by estimating cointegrating vectors for multiple equations. We present a spatial general equilibrium model for Israel in which the regional state variables include house prices, housing starts and completions, wages, employment, population and capital. Whereas housing variables are globally cointegrated, wages, employment and capital or locally cointegrated. Confidence intervals for the parameters of cointegrating vectors are calculated using the bootstrap method described in Chap. 7.

Original languageAmerican English
Title of host publicationAdvances in Spatial Science
PublisherSpringer International Publishing
Number of pages36
StatePublished - 2019

Publication series

NameAdvances in Spatial Science
ISSN (Print)1430-9602
ISSN (Electronic)2197-9375

Bibliographical note

Publisher Copyright:
© 2019, Springer Nature Switzerland AG.


Dive into the research topics of 'Cointegration in Non-Stationary Spatial Panel Data'. Together they form a unique fingerprint.

Cite this