kullnagar.boot {DCluster} | R Documentation |
Generate bootstrap replicates of Kulldorff and Nagarwalla's statistic, by calling functions boot and kullnagar.stat.
kullnagar.boot is used when using non-parametric bootstrap to estimate the distribution of the statistic.
kullnagar.pboot is used when performing parametric bootstrap.
kullnagar.boot(data, i, ...) kullnagar.pboot(...)
data |
A dataframe with the data as explained in DCluster. |
i |
Permutation created in non-parametric bootstrap. |
... |
Additional arguments passed to the functions. |
Both functions return the value of the statistic.
Kulldorff, Martin and Nagarwalla, Neville (1995). Spatial Disease Clusters: Detection and Inference. Statistics in Medicine 14, 799-810.
DCluster, boot, kullnagar, kullnagar.stat, kn.iscluster
library(boot) library(spdep) data(nc.sids) sids<-data.frame(Observed=nc.sids$SID74) sids<-cbind(sids, Expected=nc.sids$BIR74*sum(nc.sids$SID74)/sum(nc.sids$BIR74)) sids<-cbind(sids, Population=nc.sids$BIR74, x=nc.sids$x, y=nc.sids$y) niter<-100 #Permutation model kn.perboot<-boot(sids, statistic=kullnagar.boot, R=niter, fractpop=.2) plot(kn.perboot)#Display results #Multinomial model kn.mboot<-boot(sids, statistic=kullnagar.pboot, sim="parametric", ran.gen=multinom.sim, R=niter, fractpop=.2) plot(kn.mboot)#Display results #Poisson model kn.pboot<-boot(sids, statistic=kullnagar.pboot, sim="parametric", ran.gen=poisson.sim, R=niter, fractpop=.2) plot(kn.pboot)#Display results #Poisson-Gamma model kn.pgboot<-boot(sids, statistic=kullnagar.pboot, sim="parametric", ran.gen=negbin.sim, R=niter, fractpop=.2) plot(kn.pgboot)#Display results