McSpatial (version 1.1.1)

subnp: Nonparametric Employment Subcenter Identification

Description

Subcenter identification based on significant residuals from LWR employment density estimates

Usage

subnp(ydens, long, lat, window = 0.5, pval = 0.10)

Arguments

ydens

The dependent variable. Generally employment density or its natural log.

long

Longitude variable, in degrees

lat

Latitude variable, in degrees

window

The window size for the nonparametric regression. Default: .50

pval

Significance level for identifying significant residuals. Default: .10, corresponding to a t-value of 1.64.

Value

• subobs
  subobs = 1 if observation is identified as being part of a subcenter.

References

McMillen, Daniel P., "Nonparametric Employment Subcenter Identification," Journal of Urban Economics 50 (2001), 448-473. McMillen, Daniel P., "Employment Densities, Spatial Autocorrelation, and Subcenters in Large Metropolitan Areas," Journal of Regional Science 44 (2004), 225-243. McMillen, Daniel P. and Stefani C. Smith, "The Number of Subcenters in Large Urban Areas," Journal of Urban Economics 53 (May 2003), 321-338.

See Also

subgs

Examples

library(maptools)
cmap <- readShapePoly(system.file("maps/CookCensusTracts.shp",
  package="McSpatial"))
cmap$acres <- cmap$AREA*640
lmat <- coordinates(cmap)
fit <- geodistance(longvar=lmat[,1], latvar=lmat[,2],  
  lotarget=-87.627800, latarget=41.881998)
cmap$dcbd <- fit$dist
dnorth <- fit$dnorth
deast <- fit$deast

cmap$empdens <- ifelse(cmap$dcbd<=10, exp(log(50) - .16*cmap$dcbd), 10)
dsub1 <- sqrt((dnorth-10)^2 + (deast+10)^2)
dsub2 <- sqrt((dnorth+15)^2 + (deast+5)^2)
cmap$empdens <- cmap$empdens + ifelse(dsub1<=3, exp(log(20) - .19*dsub1), 0)
cmap$empdens <- cmap$empdens + ifelse(dsub2<=3, exp(log(20) - .19*dsub2), 0)
cmap$emp <- round(cmap$empdens*cmap$acres)
fit <- subnp(cmap$empdens,lmat[,1],lmat[,2])
cmap$subobs <- fit$subobs

mapplot(cmap,"subobs")