View article

This paper proposes a model for, and investigates the consequences of, strong spa% tial dependence in economic variables. Our approach and findings echo those of the corresponding lunit rootmtime series literature: We suggest a model for spatial I (1) processes, and establish a functional central limit theorem that justifies a large sam% ple Gaussian process approximation for such processes. We further generalize the I (1) model to a spatial llocal% to% unitymmodel that exhibits weak mean reversion. We char% acterize the large sample behavior of regression inference with spatial I (1) variables, and establish that spurious regression is as much a problem with spatial I (1) data as it is with time series I (1) data. We develop asymptotically valid spatial unit root tests, stationarity tests, and inference methods for the local% to% unity parameter. Finally, we consider strategies for valid inference in regressions with persistent (I (1) or local% to% unity) spatial data, such as spatial analogues of first% differencing transformations.