[R] pruning data

[William Dunlap]

> -----Original Message-----
> From: r-help-bounces at r-project.org 
> [mailto:r-help-bounces at r-project.org] On Behalf Of Payam Minoofar
> Sent: Wednesday, September 02, 2009 10:09 AM
> To: r-help at r-project.org
> Subject: [R] pruning data
> 
> Hello everyone,
> 
> I am trying to prune a data frame for partial least squares 
> analysis.  
> I need to delete an entire row if one cell in the row contains a NA.
> 
> Presently, I am running a loop that is supposed to extract the rows  
> that are full of numbers into a second data frame and skips the rows  
> that contain a single NA value.
> 
> I want to know if there is a simple way to determine if a row (about  
> 20 columns) contains a single NA value without running a loop that  
> checks each individual cell.

?na.omit

E.g.,

> x<-data.frame(one=c(NA,1,2,3), two=c("He","She","It",NA))
> x
  one  two
1  NA   He
2   1  She
3   2   It
4   3 <NA>
> na.omit(x)
  one two
2   1 She
3   2  It

Many modelling functions have an na.action argument that
takes a function like na.omit or na.fail so you don't have to keep
an NA-less version of your dataset around.
 
> 
> Thanks in advance.
> 
> __________________
> Payam Minoofar, Ph.D.
> Scientist
> Meissner Filtration Products
> 4181 Calle Tesoro
> Camarillo, CA 93012
> +1 805 388 9911 ext. 159
> +1 805 388 5948 fax
> payam.minoofar at meissner.com
>