Skip to content

methods on missing data

4 messages · Maciej.Hoffman-Wecker@evotecoai.com, Thomas Lumley, Achim Zeileis +1 more

Original
Maciej.Hoffman-Wecker@evotecoai.com
# 
Hello everybody!

I have 2 >issues< concerning methods applied to missing data.
I think they're bugs, but who knows.

1. var(NA) returns

     Error in var(NA) : missing observations in cov/cor

instead of NA. I expanded the summary-function to my.summary
including SDev, in order to use it with tapply, which crashes
in case of groups with no valid data.

2. is a similar problem. I use tapply with hist. On groups
without valid data it crashes, like hist(as.numeric(c(NA)))
returning

     Error in pretty(rx, n = nnb, min.n = 1) : x must be numeric

hist(as.numeric(c(NA)),breaks=c(1,2)) works, as it does not use "pretty".

I know these are very special cases, but unfortunately they kill the whole
process.

I can figure out some personal solution, but I hope for a professional one.

Hope this is kind of helpful.

Maciej



sysname "Linux"
release "2.4.4"
version "#1 SMP Wed May 2 20:47:43 CEST 2001"
nodename "lxnf"
machine "i686"
login "unknown"
user "hoffman"

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
Thomas Lumley
#
On Wed, 6 Jun 2001 Maciej.Hoffman-Wecker at evotecoai.com wrote:

            

      
      
      
    

        
var() already has a na.rm option that does what you want
R> var(NA,na.rm=TRUE)
[1] NA

            

      
      
      
    

        
try() is a general solution to the problem of functions that must
sometimes be applied to invalid data without causing an error.  Unlike the
var(NA) case, hist(NA) can't really do anything useful so try() is a
reasonable solution.

	-thomas

Thomas Lumley			Asst. Professor, Biostatistics
tlumley at u.washington.edu	University of Washington, Seattle

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

  
  


  


      

    

    

      
      

        


  

        

      Achim Zeileis
    

    

      
      #
    

  


  

      
Maciej.Hoffman-Wecker at evotecoai.com wrote:

            

      
      
      
    

        
I don't think so :-)

            

      
      
      
    

        
Because you didn't tell var() that it should remove missing values, it
warns you that there are missing values. If you want to remove them the
following gives the answer you expected.

            

      
      
      
    

        
[1] NA

            

      
      
      
    

        
Maybe you should check before if your vector contains only NA's, e.g.,
using is.na()
 
Hope this helps
Achim


---------------------------
Achim Zeileis
Institut f?r Statistik
Technische Universit?t Wien
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

  
  


  


      

    

    

      
      

        


  

        

      Brian Ripley
    

    

      
      #
    

  


  

      
On Wed, 6 Jun 2001 Maciej.Hoffman-Wecker at evotecoai.com wrote:

        

            

      
      
      
    

        
It works exactly as documented: see ?var.  Did you intend
to use

            

      
      
      
    

        
[1] NA

?  I suspect so, so the bug does not seem to be in R....