[R] Mixture of Distributions

[David Winsemius]

David Winsemius <dwinsemius at comcast.net> wrote in
news:Xns9A56ED77CB8C4dNOTwinscomcast at 80.91.229.13: 
> 
> #-----toy code-------
> rf1 <- function(n){rnorm(n,0,1)} ## normal distribution with mean=0
> rf2 <- function(n){rnorm(n,4,1)} ## normal distribution with mean=4
> rf3 <- function(n){rnorm(n,9,1)} ## normal distribution with mean=9
> S0 <- cbind(rf1(10),rf2(10),rf3(10))
> p1 <- 0.2; p2 <- 0.3; p3 <-0.5
> set.seed<-37

That doesn't do anything useful. 
set.seed(123) 
...will result in an ix that has at least one of each of  "1", "2", and 
"3", .

> set.seed(5)
> ix <- sample(c(1,2,3),10,prob=c(p1,p2,p3),replace=TRUE)
> ix
 [1] 3 2 1 3 3 2 2 1 1 3

-- 
David Winsemius