Skip to content

R CMD build error

6 messages · Christian Hennig, Rangesh Kunnavakkam, Jean Eid +3 more

Original
Christian Hennig
# 
Dear list,

I try to update the prabclus package.
R CMD check works nicely, no warnings, good results in all tests.
However, building the package fails:

ginkgo:/disk5/home/chrish/RAusw/libsrc R CMD build prabclus
* checking for file 'prabclus/DESCRIPTION' ... OK
* preparing 'prabclus':
* checking whether 'INDEX' is up-to-date ... OK
* removing junk files
* building 'prabclus_2.0-1.tar.gz'
tar: prabclus/man/.#prabinit.Rd: Cannot stat: No such file or directory
tar: Error exit delayed from previous errors

Does anybody have an idea why tar is looking for
prabclus/man/.#prabinit.Rd (which indeed doesn't exist, but why should
it?)

Thank you,
Christian

*** NEW ADDRESS! ***
Christian Hennig
University College London, Department of Statistical Science
Gower St., London WC1E 6BT, phone +44 207 679 1698
chrish at stats.ucl.ac.uk, www.homepages.ucl.ac.uk/~ucakche
Rangesh Kunnavakkam
# 
I  have a large vector of around 12597 elements and I wish to calculate
p-value for each element using a formula of something like:
                        p-value= 1- exp^(kexp^(-labda))
I was wondering someone could give some ideas how to implement for each element.
thankyou very much
Rangesh.K
Jean Eid
# 
if labda is the elements of the vector and you know what kexp is , you can
use apply

apply(your_vector, 1, function(x) 1- exp^(kexp^(-x)))


HTH

Jean
On Wed, 3 Aug 2005, Rangesh Kunnavakkam wrote:

            

      
      
      
    

        

      
      
    

  
  


  


      

    

    

      
      

        


  

        

      Sundar Dorai-Raj
    

    

      
      #
    

  


  

      
Rangesh Kunnavakkam wrote:

            

      
      
      
    

        
Each element of what? lambda (sic), k? What are these? vectors, scalars? 
R is vectorised, so you should be able to do following, assuming lambda 
is a vector and k is a scalar, for example:

p <- 1 - exp(k * exp(-lambda))

(I am assuming I parsed your formula correctly. If not, please do clarify.)

If both k and lambda are both vectors then the above still applies as 
long as they are the same length (including length 1). If they differ in 
length, beware of R's recycling rules.

HTH,

--sundar

P.S. Have you read the posting guide and subsequent literature, 
especially, an Introduction to R? Because you used "exp^(...)" and 
"kexp" suggests, no.

  
  


  


      

    

    

      
      

        


  

        

      Haibo Huang
    

    

      
      #
    

  


  

      
Hi,

I tried to do a logistic regression with polr(MASS). I
thought I already converted the response to factor,
but obvious I was wrong. Could anyone tell me what I
did wrong and how to correct it? Thank you very much!

            

      
      
      
    

        
as.integer(factor(Lease$EarlyTermination))

            

      
      
      
    

        
RegA=polr(ID~1+MSA+SIC.Code+TenantOption+LLOption+TOExercised,

+  data=Lease, method=c("logistic"))
Error in polr(ID ~ 1 + MSA + SIC.Code + TenantOption +
LLOption + TOExercised,  : 
        response must be a factor

            

      
      
      
    

        
Best,
Ed.

        
--- Jean Eid <jeaneid at chass.utoronto.ca> wrote:

        

            

      
      
      
    

        

      
      
    

        

      
      
    

  
  


  


      

    

    

      
      

        


  

        

      Francisco J. Zagmutt
    

    

      
      #
    

  


  

      
Why are you adding as.integer before the factor statement?  You are forcing 
the variable to be an integer even tough you are passing factor within the 
statement.  Try

Lease$ID <-factor(Lease$EarlyTermination)

Cheers

Francisco