Molly Gutilla - Creating neighbors in spatial data

Prepare packages and data

# load libraries
library(maptools)

## Warning: package 'maptools' was built under R version 2.15.3

## Loading required package: foreign

## Loading required package: sp

## Loading required package: grid

## Loading required package: lattice

## Checking rgeos availability: FALSE Note: when rgeos is not available,
## polygon geometry computations in maptools depend on gpclib, which has a
## restricted licence. It is disabled by default; to enable gpclib, type
## gpclibPermit()

library(RANN)
library(spdep)

## Warning: package 'spdep' was built under R version 2.15.3

## Loading required package: boot

## Attaching package: 'boot'

## The following object(s) are masked from 'package:lattice':
## 
## melanoma

## Loading required package: Matrix

## Loading required package: MASS

## Loading required package: nlme

## Loading required package: deldir

## deldir 0.0-21

## Loading required package: coda

## Warning: package 'coda' was built under R version 2.15.3

## Loading required package: splines

poverty <- readShapePoly(file.choose())

class(poverty)

## [1] "SpatialPolygonsDataFrame"
## attr(,"package")
## [1] "sp"

Creating neighbors

# create contiguity matrices
poverty_nbq <- poly2nb(poverty)  #default is queen, creates queen
poverty_nbr <- poly2nb(poverty, queen = FALSE)  #create rook

# choose the k nearest points as neighbors
coords <- coordinates(poverty)
FIPS <- row.names(as(poverty, "data.frame"))  #define the fips code so we can use k nearest neighbors

poverty_kn1 <- knn2nb(knearneigh(coords, k = 1), row.names = FIPS)  #create k nearest neighbors

poverty_kn2 <- knn2nb(knearneigh(coords, k = 2), row.names = FIPS)

poverty_kn4 <- knn2nb(knearneigh(coords, k = 4), row.names = FIPS)

# create distance based neighbors

# define distance
dist <- unlist(nbdists(poverty_kn1, coords))
summary(dist)  #max is 1.466

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##  0.0083  0.2800  0.3450  0.3720  0.4210  1.4700

# create max distance
max_k1 <- max(dist)

# create distance based matrix
poverty_kd1 <- dnearneigh(coords, d1 = 0, d2 = 0.75 * max_k1, row.names = FIPS)

poverty_kd2 <- dnearneigh(coords, d1 = 0, d2 = 1 * max_k1, row.names = FIPS)

poverty_kd3 <- dnearneigh(coords, d1 = 0, d2 = 1.5 * max_k1, row.names = FIPS)

# create weights matrices (row standard and binary), also exclude island
poverty_nbq_w <- nb2listw(poverty_nbq, zero.policy = T)

# use binary weights
poverty_nbq_wb <- nb2listw(poverty_nbq, style = "B", zero.policy = T)

# run moran's I test, using created weights
moran.test(poverty$PROPOV, listw = poverty_nbq_w, zero.policy = T)

## 
##  Moran's I test under randomisation
## 
## data:  poverty$PROPOV  
## weights: poverty_nbq_w  
##  
## Moran I statistic standard deviate = 57.52, p-value < 2.2e-16
## alternative hypothesis: greater 
## sample estimates:
## Moran I statistic       Expectation          Variance 
##         0.6194394        -0.0003224         0.0001161