> #######################################################################
> 
> # Nonparametric Bootstrap using bootstrap(), emp, bt, bca
> 
> # function to compute the correlation and Fisher's z-transformation
> 
> cor2 <- function(X){
+ 	z1 <- X[,1]
+ 	z2 <- X[,2]
+ 	a <- cor(z1,z2)
+ 	b <- 0.50*(log((1+a)/(1-a)) - log((1+r.samp)/(1-r.samp)))/sqrt(n-3)
+ 	return(c(a,b))
+ }
> 
> r.boot <- bootstrap(X, cor2)
Forming replications  1  to  100 
Forming replications  101  to  200 
Forming replications  201  to  300 
Forming replications  301  to  400 
Forming replications  401  to  500 
Forming replications  501  to  600 
Forming replications  601  to  700 
Forming replications  701  to  800 
Forming replications  801  to  900 
Forming replications  901  to  1000 

> 
> plot(r.boot)
> 
> print(r.boot)
Call:
bootstrap(data = X, statistic = cor2)

Number of Replications: 1000 

Summary Statistics:
      Observed      Bias     Mean      SE 
cor21   0.8766 -0.001532 0.875098 0.04647
cor22   0.0000  0.005074 0.005074 0.03748
> 
> secor <- r.boot$estimate[1,3]	# location of the SE in the bootstrap data.frame
> secor
[1] 0.0465
> 
> r.boot.emp <- limits.emp(r.boot)
> r.boot.emp
         2.5%      5%    95%  97.5% 
cor21  0.7666  0.7901 0.9362 0.9413
cor22 -0.0672 -0.0557 0.0665 0.0747
> 
> finv025 <- r.boot.emp[2,1]
> finv025
[1] -0.0672
> 
> finv975 <- r.boot.emp[2,4]
> finv975
[1] 0.0747
> 
> r.btci <- c(r.samp - (secor*finv95), r.samp - (secor*finv05))
> r.btci
[1] 0.824 0.962
> 
> r.boot.bca <- limits.bca(r.boot, details = T)
> r.boot.bca
$limits:
         2.5%      5%    95%  97.5% 
cor21  0.7231  0.7518 0.9273 0.9344
cor22 -0.0876 -0.0755 0.0517 0.0614

$emp.probs:
         2.5%     5%   95% 97.5% 
cor21 0.00573 0.0177 0.900 0.937
cor22 0.00417 0.0147 0.894 0.932

$z0:
  cor21  cor22 
 -0.103 -0.103

$acceleration:
   cor21   cor22 
 -0.0724 -0.0902

$group.size:
[1] 1

> 
>