get.pi.bootstrap {IDSpatialStats}

Bootstrap get.pi values.

Description

Runs get.pi on multiple bootstraps of the data. Is formulated such that the relationships between points and themselves will not be calculated.

Usage

get.pi.bootstrap(
  posmat,
  fun,
  r = 1,
  r.low = rep(0, length(r)),
  boot.iter = 500,
  data.frame = TRUE
)

Arguments

posmat

a matrix with columns type, x and y

fun

the function to decide relationships

r

the series of spatial distances we are interested in

r.low

the low end of each range. 0 by default

boot.iter

the number of bootstrap iterations

data.frame

logical indicating whether to return results as a data frame (default = TRUE)

Value

pi values for all the distances we looked at

Note

In each bootstrap iteration N observations are drawn from the existing data with replacement. To avoid errors in inference resulting from the same observatin being compared with itself in the bootstrapped data set, original indices are perserved, and pairs of points in the bootstrapped dataset with the same original index are ignored.

Author(s)

Justin Lessler and Henrik Salje

See Also

Other get.pi: get.pi(), get.pi.ci(), get.pi.permute(), get.pi.typed(), get.pi.typed.bootstrap(), get.pi.typed.permute()

Examples

#compare normally distributed with uniform points
x<-cbind(1,runif(100,-100,100), runif(100,-100,100))
x<-rbind(x, cbind(2,rnorm(100,0,20), rnorm(100,0,20)))
colnames(x) <- c("type","x","y")

fun<-function(a,b) {
    if(a[1]!=2) return(3)
    if (b[1]==2) return(1)
    return(2)
}

r.max<-seq(10,100,10)
r.min<-seq(0,90,10)
r.mid <- (r.max+r.min)/2


pi<-get.pi(x,fun,r=r.max,r.low=r.min)
pi.boot<-get.pi.bootstrap(x,fun,r=r.max,r.low=r.min,boot.iter=100)

pi.ci<-apply(pi.boot[,-(1:2)],1,quantile,probs=c(0.25,0.75))

plot(r.mid, pi$pi , type="l")
lines(r.mid, pi.ci[1,] , lty=2)
lines(r.mid, pi.ci[2,] , lty=2)