R/lplotPV.R

In musto101/wilcox_R: What the package does (short line)

lplotPV <-
function(x,y, span = 0.75, xout = FALSE,pr=TRUE,
    outfun = out,nboot=1000,SEED=TRUE,plotit=TRUE,pyhat = FALSE, expand = 0.5, low.span = 2/3,
    varfun = pbvar, cor.op = FALSE, cor.fun = pbcor, scale = FALSE,
    xlab = "X", ylab = "Y", zlab = "", theta = 50, phi = 25,
    family = "gaussian", duplicate = "error", pc = "*", ticktype = "simple",...){
#
# Compute a p-value based on the Strength of Association estimated via lplot
# If significant, conclude there is dependence.
#
if(SEED)set.seed(2)
x=as.matrix(x)
if(ncol(x)==2 && !scale){
if(pr){
print("scale=F is specified.")
print("If there is dependence, might use scale=T")
}}
vals=NA
nv=ncol(x)
m=elimna(cbind(x,y))
x<-m[,1:nv]
y<-m[,nv+1]
if(xout){
flag<-outfun(x,plotit=FALSE,...)$keep
m<-m[flag,]
x<-m[,1:nv]
y<-m[,nv+1]
}
x=as.matrix(x)
est=lplot(x,y,span=span,plotit=plotit,pr=FALSE, pyhat = pyhat,
    outfun = outfun, expand = expand, low.span = low.span,
    varfun = varfun, cor.op =cor.op, cor.fun = cor.fun, scale = scale,
    xlab = xlab, ylab = ylab, zlab =zlab, theta =theta, phi = phi,
    family = family, duplicate = duplicate, pc = pc, ticktype = ticktype,...)
n=nrow(x)
data1<-matrix(sample(n,size=n*nboot,replace=TRUE),nrow=nboot)
data2<-matrix(sample(n,size=n*nboot,replace=TRUE),nrow=nboot)
for(i in 1:nboot){
vals[i]=lplot(x[data1[i,],],y[data2[i,]],plotit=FALSE,pr=FALSE)$Strength.Assoc
}
p=mean(est$Strength<vals)
list(p.value=p,Strength.Assoc=est$Strength.Assoc,Explanatory.power=est$Explanatory.power,yhat.values=est$yhat.values)
}