InverseCDFNormal.R

Inverse CDF Method to Generate Random Variable cc @mathsorosThe following R code is used to generate certain number of random variables under N(0,1) distribution. The main function is ?sampler? and it take an integer as input.

## Main function ##
sampler <- function(n){
  rv <- NULL    ## Initialization ##
  for (i in 1:n){    
  Y <- runif(1,min=0,max=1)
  ## Change this to generate rv's under other distribution ##
  X <- qnorm(Y,mean=0,sd=1)       
  rv <- c(rv,X)    
  }   
  return(rv) 
}
## Generate 10000 N(0,1) random variables ##
genrv <- sampler(10000)
## Plot histogram of samples ##
hist(genrv,30,prob=TRUE,xlab='X',          
     main='Normal random samples generated by inverse CDF')
## Compare with N(0,1) distribution
xx <- seq(-3,3,length.out=10000) 
lines(xx,dnorm(xx),type="l",col="blue")