[R] question about a program

Jim Holtman jholtman at gmail.com
Wed Jun 23 15:05:30 CEST 2010


use Rprof to determine where your function is spending time.

What is the problem you are trying to solve?

Sent from my iPhone.

On Jun 23, 2010, at 5:21, li li <hannah.hlx at gmail.com> wrote:

> Dear all,
>   I have the following program for a multiple comparison procedure.
> There are two functions for the two steps. First step is to  
> calculate the
> critical values,
> while the second step is the actual procedure [see below: program  
> with two
> functions].
>   This  work fine. However, However I want to put them into one  
> function
> for the convenience
> of later use [see below: program with one function]. Some how the big
> function works extremely
> slow.  For example I chose
> m <- 10
> rho <- 0.1
> k <- 2
> alpha <- 0.05
> pvaluessort <- sort(1-pnorm(rnorm(10))
>
> Is there anything that I did wrong?
> Thank you!
>                    Hannah
>
>
> ######Program with two functions############
> ## first step
> library(mvtnorm)
> cc_f <- function(m, rho, k, alpha) {
>
> cc_z <- numeric(m)
> var <- matrix(c(1,rho,rho,1), nrow=2, ncol=2, byrow=T)
> for (i in 1:m){
>   if (i <= k) {cc_z[i] <- qmvnorm((k*(k-1))/(m*(m-1))*alpha,  
> tail="upper",
> sigma=var)$quantile} else
>               {cc_z[i] <- qmvnorm((k*(k-1))/((m-i+k)*(m-i+k-1))*alpha,
> tail="upper", sigma=var)$quantile}
>               }
> cc <- 1-pnorm(cc_z)
> return(cc)
>                             }
> ## second step
> pair_kFWER=function(m,crit_cons,pvaluessort){
> k=0
> while((k <m)&&(crit_cons[m-k] < pvaluessort[m-k])){
> k=k+1
> }
> return(m-k)
> }
> ###########################################
> ##############Program with one function ##############
>
> pair_kFWER=function(m,alpha,rho,k,pvaluessort){
> library(mvtnorm)
> cc_z <- numeric(m)
> var <- matrix(c(1,rho,rho,1), nrow=2, ncol=2, byrow=T)
> for (i in 1:m){
>   if (i <= k) {cc_z[i] <- qmvnorm((k*(k-1))/(m*(m-1))*alpha,  
> tail="upper",
> sigma=var)$quantile} else
>               {cc_z[i] <- qmvnorm((k*(k-1))/((m-i+k)*(m-i+k-1))*alpha,
> tail="upper", sigma=var)$quantile}
>               }
> crit_cons <- 1-pnorm(cc_z)
>
> k=0
> while((k <m)&&(crit_cons[m-k] < pvaluessort[m-k])){
> k=k+1
> }
> return(m-k)
> }
> #####################################################
>
>    [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.



More information about the R-help mailing list