Raptis V S,Vaughan R A,Wright G G.The effect of scal-ing on land cover classification from satellite date.Com-puters and Geosciences,2003,29(6):705~714.
Latifovic R,Olthof I.Accuracy assessment using sub-pixelfractional error matrices of global land cover products de-rived from satellite data.Remote Sensing of Environment,2004,90(2):153~165.
[4]
Ju J,Kolaczyk E D,Gopal S.Gaussian mixture discriminateanalysis and sub-pixel land cover characterization in re-mote sensing.Remote Sensing of Environment,2003,84(4):550~560
Giles M F.GIS:stressing the geographical.Progress inPhysical Geography,2004,281:152~158.
[9]
Heegaard E.The outer border and central border forspecies-environmental relationships estimated by nonpara-metric generalized additive models.Ecological Modelling,2002,157(2~3):131~139.
[10]
Guisan A,Edwards J T C,Hastie T.Generalised linear andgeneralised additive models in studies of species distribu-tions:setting the scene.Ecological Modelling,2002,157(2~3):89~100.
[11]
Austin M P.Spatial prediction of species distribution:aninterface between ecological theory and statistical model-ing.Ecological Modelling,2002,157(2~3):101~118.
[12]
Miller J,Franklin J.Modelling the distribution of four veg-etation alliances using generalised linear models and clas-sification tress with spatial dependence.Ecological Mod-elling,2002,157(2~3):227~247.
[13]
King A.Translating models across scales in the landscape.In:Turner M G,Garder R H,et al.Quanitative Methods inLandscape Ecology.New York:Springer Verlag,1991,479~517.
[14]
Leung Y,Mei C L,Zhang WX.Statistical test for spatialnonstationarity based on the geographically weighted re-gression model.Environment and Planning A,2000a,32:9~32.
[15]
Leung Y,Mei C L,Zhang W X.Testing for spatial autocor-relation among the residuals of the geographically weightedregression.Environment and Planning A,2000b,32:871~890.
[16]
Brunsdon C,McClatchey J,Unwin D J.Spatial variations inthe average rainfall altitude relationship in Great Britain:an approach using geographically weighted regression.In-ternational Journal of Climatology,2001,21:455~66.
[17]
Paez A,Uchida T,Miyamoto K.A general framework for es-timation and inference of geographically weighted regres-sion models:1.Location-specific kernel bandwidths and atest for local heterogeneity.Environment and Planning A,2002a,34:733~754.
[18]
Paez A,Uchida T,Miyamoto K.A general framework for es-timation and inference of geographically weighted regres-sion models:2.Spatial association and model specificationtests.Environment and Planning A,2002b,34:883~904.
[19]
Nakaya T.Local spatial interaction modelling based on thegeographically weighted regression approach.In Thomas R,Boots B,Okabe A.Modelling Geographical Aystems:Sta-tistical and Computational Applications.Dordrecht:Kluwer,2003:45~69.
[20]
Atkinson P M,German S E,Sear D A,et al.Exploring therelations between riverbank erosion and geomorphologicalcontrols using geographically weighted logistic regression.Geographical Analysis,2003,35:58~82.
[21]
Yefang Huang,Yee Leung.Analysing regional industriali-sation in Jiangsu province using geographically weightedregression.Geographical Systems,2002,4(2):233~249.
[22]
Osborne P E,Suarez-Seoane S.Should data be partitioned·68·spatially before building large-scale distribution models?.Ecological Modelling,2002,157(2~3):249~259.