Modeling Spatial Autocorrelation in Spatial Interaction Data: A Comparison of Spatial Econometric and Spatial Filtering Specifications
Manfréd M. Fischer, Daniel A. Griffith
Abstract
Open-access reader
Manfréd M. Fischer, Daniel A. Griffith
Abstract
Open-access reader
The need to account for spatial autocorrelation is well known in spatial analysis. Many spatial statistics and spatial econometric texts detail the way spatial autocorrelation can be identified and modelled in the case of object and field data. The literature on spatial autocorrelation is much less developed in the case of spatial interaction data. The focus of interest in this paper is on the problem of spatial autocorrelation in a spatial interaction context. The paper aims to illustrate that eigenfunction-based spatial filtering offers a powerful methodology that can efficiently account for spatial autocorrelation effects within a Poisson spatial interaction model context that serves the purpose to identify and measure spatial separation effects to interregional knowledge spillovers as captured by patent citations among high-technology-firms in Europe.
OpenAlex reports 9 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
The need to account for spatial autocorrelation is well known in spatial analysis. Many spatial statistics and spatial econometric texts detail the way spatial autocorrelation can be identified and modelled in the case of object and field data. The literature on spatial autocorrelation is much less developed in the case of spatial interaction data. The focus of interest in this paper is on the problem of spatial autocorrelation in a spatial interaction context. The paper aims to illustrate that eigenfunction-based spatial filtering offers a powerful methodology that can efficiently account for spatial autocorrelation effects within a Poisson spatial interaction model context that serves the purpose to identify and measure spatial separation effects to interregional knowledge spillovers as captured by patent citations among high-technology-firms in Europe.
Key concepts: Spatial analysis, Spatial contextual awareness, Autocorrelation, Spatial econometrics, Context (archaeology), Spatial dependence, Econometrics, Computer science