boot: Non parametric bootstrap for a fitted correlation.

In jeroenooms/JJcorr: Calculates polychorical correlations for several copula families.

Description

This function performs a non parametric bootstrap for a fitted correlation. The argument should be an object returned by PC, empPC or spearmanR.

Usage

	boot(model, size)

Arguments

model	a correlation object as returned by functions PC, empPC or spearmanR.
size	the number of resamples

Value

model	copy of the object argument model
r	vector with correlation (r) values for every sample
theta	only for PC models: vector of fitted theta values for the Copula distribution
loss	only for PC models: vector of loss values for the fitted Copula distribution

Examples

#load contingency table
data(table4);

#fit a copula correlation
myOC <- PC(table4,"gauss","MH");
bs1 <- boot(myOC, size=10);
print(bs1);

#fit an empirical correlation
myEC <- empPC(table4);
bs2 <- boot(myEC);

#fit a spearman correlation
mySR <- spearmanR(table4);
bs3 <- boot(mySR);

jeroenooms/JJcorr documentation built on May 19, 2019, 6:10 a.m.