R help - Mar 2006 - ignoring objects containing NaN's in a recursive function

Hi-

General question:
I have written a function that performs a permutation test.  I use the
function "sample" to randomize my data, then I call on another
function
I wrote to analyze the randomized data.  This process repeats "n"
times-- 10, 100 or whatever.

The function applied to the randomized data creates a matrix and
sometimes the matrix contains NaN's.  I would like my function to simply
ignore any matrices that contain NaN's and continue to cycle until it
has generated "n" matrices without any NaN's.  I am aware of the
function "is.nan", but I'm not sure how to fit it in.



Specifics:
Here's the section of code that needs improvement:

for (i in 1:n){
	S[,,]= sample (Y[,,])
	dig=auto(S,H,J,T)	
	out[i,]=dig[1,]}


n= number of randomizations
S= array to store randomized data
Y= original data array
dig= the matrix returned by the function "auto"
out= matrix that collects the first row of each "dig" matrix up to
"n"

Sometimes "dig" includes NaN's.  I want the function to ignore
those
"dig"s but keep working until it has generated "n" matrices
that contain
no NaN's.  

How could I do this?

Thanks.

Jason Mills

if(any(is.nan(...)))... ##repeat

However,

1) Have you tried RSiteSearch('permutation
test',restr='function') to see
what's already there?

2) I suspect that there are better ways to do this. In particular,
generating all random whatevers at once is way more efficient than repeated
calls;

2) It is **not** a permutation test, since there is no guarantee that all
the calls to sample() generate different permutations. (depending on the
sample size, this may be nitpicking).

Cheers,

-- Bert Gunter
Genentech Non-Clinical Statistics
South San Francisco, CA
 
"The business of the statistician is to catalyze the scientific learning
process."  - George E. P. Box
 
 
> -----Original Message-----
> From: r-help-bounces at stat.math.ethz.ch 
> [mailto:r-help-bounces at stat.math.ethz.ch] On Behalf Of Mills, Jason
> Sent: Tuesday, March 14, 2006 11:00 AM
> To: r-help at stat.math.ethz.ch
> Subject: [R] ignoring objects containing NaN's in a recursive function
> 
> Hi-
> 
> General question:
> I have written a function that performs a permutation test.  I use the
> function "sample" to randomize my data, then I call on 
> another function
> I wrote to analyze the randomized data.  This process repeats "n"
> times-- 10, 100 or whatever.
> 
> The function applied to the randomized data creates a matrix and
> sometimes the matrix contains NaN's.  I would like my 
> function to simply
> ignore any matrices that contain NaN's and continue to cycle until it
> has generated "n" matrices without any NaN's.  I am aware of
the
> function "is.nan", but I'm not sure how to fit it in.
> 
> 
> 
> Specifics:
> Here's the section of code that needs improvement:
> 
> for (i in 1:n){
> 	S[,,]= sample (Y[,,])
> 	dig=auto(S,H,J,T)	
> 	out[i,]=dig[1,]}
> 
> 
> n= number of randomizations
> S= array to store randomized data
> Y= original data array
> dig= the matrix returned by the function "auto"
> out= matrix that collects the first row of each "dig" matrix up
to "n"
> 
> Sometimes "dig" includes NaN's.  I want the function to
ignore those
> "dig"s but keep working until it has generated "n"
matrices
> that contain
> no NaN's.  
> 
> How could I do this?
> 
> Thanks.
> 
> Jason Mills
> 
> ______________________________________________
> 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
>