Abstract
Although the assumption of independence among interaction flows frequently is engaged in spatial interaction modeling, in many circumstances it leads to misspecified models and incorrect inferences. An informed approach is to explicitly incorporate an assumed relationship structure among the interaction flows, and to explicitly model the network autocorrelation. This paper illustrates such an approach in the context of U.S. interstate migration flows. Behavioral assumptions, similar to those of the intervening opportunities or the competing destinations concepts, exemplify how to specify network flows that are related to particular origin–destination combinations. The stepwise incorporation of eigenvectors, which are extracted from a network link matrix, captures the network autocorrelation in a Poisson regression model specification context. Spatial autocorrelation in Poisson regression is measured by the test statistic of Jacqmin-Gadda et al. (Stat Med 16(11):1283–1297, 1997). Results show that estimated regression parameters in the spatial filtering interaction model become more intuitively interpretable.
Similar content being viewed by others
Notes
See Guldmann (1999) for a more detailed discussion in the definition of geographic space over which intervening opportunity factors may be computed.
Note that eigenvector spatial filtering treats the eigenvectors of the transformed spatial link matrix V as proxies for genuine unobserved variables.
Under “queen” type contiguity, spatial contiguity is defined as sharing a common boundary or vertex.
The 95% confidence intervals for the distance parameters from the unfiltered and spatially filtered model specifications are (−0.4680, −0.3668) and (−0.4395, −0.3729), respectively.
Note that one confounding factor that should be mentioned is the national economic growth that is occurring. Its geographic distribution focuses on warmer climates because of such regional attributes as non-union labor, relatively inexpensive and abundant available land, and lower taxes (Hanson 1998).
The 95% confidence intervals for the distance parameters from the unfiltered and spatially filtered model specifications are (−0.5491, −0.4459) and (−0.4660, −0.4002), respectively.
References
Almeida LMW, Goncalves MB (2001) A methodology to incorporate behavioral aspects in trip-distribution models with an application to estimate student flow. Environ Plan A 33(6):1125–1138
Anselin L, Griffith DA (1988) Do spatial effects really matter in regression analysis? Papers Reg Sci Assoc 65:11–34
Assuncao RM, Reis EA (1999) A new proposal to adjust Moran’s I for population density. Stat Med 18:2147–2162
Baxter M (1983) Estimation and inference in spatial interaction models. Progr Hum Geogr 7:40–59
Berglund S, Karlstrom A (1999) Identifying local spatial association in flow data. J Geogr Syst 1(3):219–236
Black WR (1992) Network autocorrelation in transport network and flow systems. Geogr Anal 24(3):207–222
Bolduc D, Dagenais MG, Gaudry MJI (1989) Spatially autocorrelated errors in origin–destination models: a new specification applied to aggregate mode choice. Transp Res Part B 23(5):361–372
Boots BN, Kanaroglou PS (1988) Incorporating the effects of spatial structure in discrete choice models of migration. J Reg Sci 28(4):495–509
Cadwallader MT (1992) Migration and residential mobility: macro and micro approaches. University of Wisconsin Press, Madison
Congdon P (1992) Aspects of general linear modelling of migration. Statistician 41:133–153
Congdon P (1993) Approaches to modelling overdispersion in the analysis of migration. Environ Plan A 25(10):1481–1510
Congdon P (2002a) A Bayesian approach to prediction using the gravity model with an application to patient flow modeling. Geogr Anal 32(3):205–224
Congdon P (2002b) A model for mental health needs and resourcing small geographic areas: a multivariate spatial perspective. Geogr Anal 34(2):169–186
Cressie N, Chan HH (1989) Spatial modeling of regional variables. J Am Stat Assoc 84(405):393–401
Cushing BJ (1987) Location-specific amenities, topography, and population migration. Ann Reg Sci 21:74–85
Davies PS, Greenwood MJ, Li H (2001) A conditional logit approach to U.S. state-to-state migration. J Reg Sci 41(2):337–360
de Jong P, Sprenger C, van Veen F (1984) On extreme values of Moran’s I and Geary’s C. Geogr Anal 16(1):17–24
Fik TJ, Amey RG, Mulligan GF (1992) Labor migration amongst hierarchically competing and intervening origins and destinations. Environ Plan A 24:1271–1290
Fischer MM, Griffith DA (2008) Modeling spatial autocorrelation in spatial interaction data: an application to patent citation data in the European Union. J Reg Sci (forthcoming)
Flowerdew R (1991) Poisson regression modelling of migration. In: Stillwell J, Congdon P (eds) Migration models: macro and micro approaches. Belhaven, New York, pp 92–112
Flowerdew R, Aitkin M (1982) A method of fitting the gravity model based on the Poisson distribution. J Reg Sci 22(2):191–202
Flowerdew R, Lovett A (1988) Fitting constrained Poisson regression models to inter-urban migration. Geogr Anal 20(4):297–307
Fotheringham AS (1983) A new set of spatial-interaction models: the theory of competing destinations. Environ Plan A 15(1):15–36
Fotheringham AS (1986) Modelling hierarchical destination choice. Environ Plan A 18(3):401–418
Fotheringham AS (1991) Migration and spatial structure: the development of the competing destination model. In: Stillwell J, Congdon P (eds) Migration models: macro and micro approaches. Belhaven, New York, pp 57–72
Fotheringham AS, O’Kelly ME (1989) Spatial interaction models: formulations and applications. Kluwer Academic Publishers, Dordrecht
Galle OR, Taeuber KE (1966) Metropolitan migration and intervening opportunities. Am Sociol Rev 31(1):5–13
Getis A (1990) Screening for spatial dependence in regression analysis. Papers Reg Sci Assoc 69:69–81
Getis A (1995) Spatial filtering in a regression framework: examples using data on urban crime, regional inequality, and government expenditures. In: Anselin L, Florax RJGM (eds) New directions in spatial econometrics. Springer, Berlin, pp 172–188
Getis A, Griffith DA (2002) Comparative spatial filtering in regression analysis. Geogr Anal 34(2):130–140
Gleave D, Cordey-Hayes M (1977) Migration dynamics and labor market turnover. Prog Plan 8(1):1–95
Greenwood MJ (1985) Human migration: theory, models, and empirical studies. J Reg Sci 25(4):521–544
Griffith DA (1996) Spatial autocorrelation and eigenfunctions of the geographic weights matrix accompanying geo-reference data. Can Geogr 40(4):351–367
Griffith DA (2000) A linear regression solution to the spatial autocorrelation problem. J Geogr Syst 2(2):141–156
Griffith DA (2002) A spatial filtering specification of the auto-Poisson model. Stat Probab Lett 58(3):245–251
Griffith DA (2003) Spatial autocorrelation and spatial filtering: gaining understanding through theory and scientific visualization. Springer, Berlin
Griffith DA (2004) A spatial filtering specification for the autologistic model. Environ Plan A 36(10):1791–1811
Griffith DA (2006) Hidden negative spatial autocorrelation. J Geogr Syst 8(4):335–355
Griffith DA (2007) Spatial structure and spatial interaction: 25 years later. Rev Reg Stud 37(1):28–38
Guldmann JM (1999) Competing destinations and intervening opportunities interaction models of inter-city telecommunication. Papers Reg Sci 78(2):179–194
Haining R (1991) Bivariate correlation and spatial data. Geogr Anal 23(3):210–227
Haining R (2003) Spatial data analysis: theory and practice. Cambridge University Press, Cambridge
Hanson GH (1998) North American economic integration and industry location. Oxford Rev Econ Policy 14:30–44
Harary F (1999) Graph theory. Perseus Books, Reading, MA
Homes JH, Haggett P (1977) Graph theory interpretation of flow matrices: a note on maximization procedures for identifying significant links. Geogr Anal 9(4):388–399
Jacqmin-Gadda H, Commenges D, Nejjari C, Dartigues JF (1997) Tests of geographical correlation with adjustment for explanatory variables: an application to Dyspnoea in the elderly. Stat Med 16(11):1283–1297
Jayet H (1990) Spatial search processes and spatial interaction. 1. Sequential search, intervening opportunities, and spatial search equilibrium. Environ Plan A 22(5):583–599
Law J, Haining R (2004) Bayesian approach to modeling binary data: the case of high-intensity crime analysis. Geogr Anal 36(3):197–216
Liao PS (2001) Contextual analysis of rural migration intention: a comparison of Taiwanese and Pennsylvania data. Int J Comp Sociol 42(5):435–460
Lin G, Zhang T (2007) Loglinear residual tests of Moran’s I autocorrelation and their applications to Kentucky breast cancer data. Geogr Anal 39(3):293–310
Myers RH, Montgomery DC, Vining GG (2002) Generalized linear models with applications in engineering and the sciences. Wiley, New York
Mulder CH, Hooimeijer P (1999) Residential relocations in the life course. In: van Wissen LJG, Dykstra PA (eds) Population issues: an interdisciplinary focus. Kluwer Academic Press, New York, pp 159–186
Nelder JA, Wedderburn RWM (1972) Generalized linear models. J R Stat Soc Ser A 35(3):370–384
Newbold KB (2005) Spatial scale, return and onward migration, and the Long-Boertlein index of repeat migration. Papers Reg Sci 84(2):281–290
Oden N (1995) Adjusting Moran’s I for population density. Stat Med 14(1):17–26
Plane DA (1984) Migration space: doubly constrained gravity model mapping of relative interstate separation. Ann Assoc Am Geogr 74(2):244–245
Prentice RL, Zhao LP (1991) Estimating equations for parameters in means and covariances of multivariate discrete and continuous responses. Biometrics 47(3):825–839
Ruiter ER (1967) Toward a better understanding of the intervening opportunities model. Transp Res 1:47–56
Shaw RP (1985) Intermetropolitan migration in Canada: changing determinants over three decades. NC Press, Toronto
Schabenberger O, Gotway CA (2005) Statistical methods for spatial data analysis. Chapman & Hall/CRC, Boca Raton
Schwartz A (1973) Interpreting the effects of distance on migration. J Polit Econ 81(5):1153–1169
Shvetsov VI, Dubov YA (1997) Expected distributions in the intervening opportunities model. Environ Plan A 29(7):1229–1241
Stouffer SA (1940) Intervening opportunities: a theory relating mobility and distance. Am Sociol Rev 5(6):845–867
Stouffer SA (1960) Intervening opportunities and competing migrants. J Reg Sci 2(1):1–26
Tiefelsdorf M (2000) Modelling spatial processes: the identification and analysis of spatial relationships in regression residuals by means of Moran’s I. Springer, Berlin
Tiefelsdorf M (2003) Misspecifications in interaction model distance decay relations: a spatial structure effect. J Geogr Syst 5(1):25–50
Tiefelsdorf M, Boots BN (1995a) The exact distribution of Moran’s I. Environ Plan A 27(6):985–999
Tiefelsdorf M, Boots BN (1995b) The specification of constrained interaction models using the SPSS Loglinear procedure. Geogr Syst 2(1):21–38
Tiefelsdorf M, Griffith DA (2007) Semiparametric filtering of spatial autocorrelation: the eigenvector approach. Environ Plan A 39(5):1193–1221
Tiefelsdorf M, Griffith DA, Boots BN (1999) A variance-stabilizing coding scheme for spatial link matrices. Environ Plan A 31(1):165–180
US Department of Commerce (2002) National Oceanic and Atmospheric Administration, Environmental Data and Information Service. State, regional, and national monthly temperature (weighted by area), 1971–2000 (and previous normals periods). National Climatic Data Center, Asheville
Wakefield JC (2003) Sensitivity analyses for ecological regression. Biometrics 59(1):9–17
Wakefield JC, Best NG, Waller LA (2000) Bayesian approaches to disease mapping. In: Elliott P, Wakefield J, Best NG, Briggs DJ (eds) Spatial epidemiology: method and application. Oxford University Press, Oxford, pp 104–127
Waldhör T (1996) The spatial autocorrelation coefficient Moran’s I under heteroscedasticity. Stat Med 15(7–9):887–892
Wilson AG (1970) Entropy in urban and regional modelling. Pion Limited, London
Acknowledgments
The author thanks Michael Tiefelsdorf for useful comments on this research, and anonymous reviewers for their comments and suggestions.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Chun, Y. Modeling network autocorrelation within migration flows by eigenvector spatial filtering. J Geogr Syst 10, 317–344 (2008). https://doi.org/10.1007/s10109-008-0068-2
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10109-008-0068-2