[R] help with writing function

[Oarabile Molaodi]

Thanks for the help Petr it works,  and all who contributed  managed 
with the  suggested function below as I needed to remove the zeros and 
the upper triangular portion of the matrix.:

IED <- function(risk) {

   n     <- length(risk)
   mrisk <- matrix( rep(risk, n), ncol=n, byrow=TRUE )
   diff  <- abs(risk - mrisk)

   diff[ lower.tri(diff) ]
 }

OR

IED2 <- function(risk){
   o <- abs( outer( risk, risk, FUN="-" ) )
   o[ lower.tri(o) ]
 }


Oarabile

Petr Pikal wrote:

>Hi
>
>or use outer
>
>x<-1:10
>x<-sample(x)
>mat<-outer(x,x,"-") # result with zeroes
>matrix(mat[!mat==0],9,9) # get rid of them
>
>will give you a matrix with columns with intended result.
>
>HTH
>Petr
>
>
>
>
>On 13 Dec 2005 at 11:53, Robert Burrows wrote:
>
>Date sent:      	Tue, 13 Dec 2005 11:53:00 -0500 (EST)
>From:           	Robert Burrows <rbb at nebiometrics.com>
>To:             	Oarabile Molaodi <oarabile at stams.strath.ac.uk>
>Copies to:      	r-help at stat.math.ethz.ch
>Subject:        	Re: [R] help with writing function
>
>  
>
>>On Tue, 13 Dec 2005, Oarabile Molaodi wrote:
>>
>>    
>>
>>>I'm trying to write a function that takes a vector of length n  and
>>>then takes the first value of the vector i.e j=1 and forms a new
>>>vector of length n (i.e replicate the first value n times). This
>>>function will then calculate the absoulte difference of the original
>>>vector and the new vector and store the results omitting the
>>>difference between the value and itself. This function should be
>>>able to repeat the procedure for each of the j's i.e j=2 to n. The
>>>results should all be stored together. Below is  what I've tried so
>>>far but it seems to work only for j=1 .
>>>
>>>Your help will be highly appreciated.
>>>IED<-function(risk){
>>>n<-length(risk)
>>>i<-c(1:n)
>>>Diff<-numeric()
>>>for(j in 1:n){
>>>relrisk<-risk
>>>relrisk[i]<-relrisk[j]
>>>Difference<-abs(risk-relrisk)
>>>Difference<-Difference[-c(1:j)]
>>>Difference<-append(Diff,Difference)
>>>return(Difference)
>>>}
>>>}
>>>      
>>>
>>How about
>>
>>"IED" <-
>>function(risk){
>>  n<-length(risk)
>>Diff<-numeric(n)
>>for(j in 1:n){
>>relrisk<-rep(risk[j],n)
>>Diff[j]<-sum(abs(risk-relrisk)[-j])
>>}
>>Diff
>>}
>>
>>-- 
>>Robert Burrows, PhD
>>New England Biometrics
>>rbb at nebiometrics.com
>>
>>______________________________________________
>>R-help at stat.math.ethz.ch mailing list
>>https://stat.ethz.ch/mailman/listinfo/r-help
>>PLEASE do read the posting guide!
>>http://www.R-project.org/posting-guide.html
>>    
>>
>
>Petr Pikal
>petr.pikal at precheza.cz
>
>  
>