Spatial cointegration and heteroscedasticity

A two-step Lagrange Multiplier test strategy has recently been suggested as a tool to reveal spatial cointegration. The present paper generalises the test procedure by incorporating control for unobserved heteroscedasticity. Using Monte Carlo simulation, the behaviour of several relevant tests for spatial cointegration and/or heteroscedasticity is investigated. The two-step test for spatial cointegration appears to be robust towards heteroscedasticity. While several tests for heteroscedasticity prove to be inconclusive under certain circumstances, a Lagrange Multiplier test for heteroscedasticity based on spatially differenced variables is shown to serve well as an indication of heteroscedasticity irrespective of cointegration status.