R help - Dec 2003 - diagnostic information ....

Paul E. Johnson wrote:
> I handed out some results from glm() and the students ask "how many 
> observations were dropped due to missing values"?
> 
> How would I know? 
> 
> In other stat programs, the results will typically include N and
> the number dropped because of missings.  Without going back to R and
> fiddling about to find the total number of rows in the dataframe,
> there is no way to tell.  Somewhat inconvenient. Do you agree?
In a word:  No.  R is ``not like other packages'' which spew out
enormous printouts including every statistic known to man and a few
known only to woman.  R basically gives you what you ask for, and
assumes you know enough to know what to ask for.  It does not
condescendingly make decisions for you and hand-cuff you into doing a
standard analysis.

It also allows infinitely versatile customization with consumate
ease since it is a beautifully designed programming language.  It
would be ***VERY*** easy to write a wrapper for glm() to include
the number of dropped observations if you want to include that
information in a ``printout''.

				cheers,

					Rolf Turner
					rolf at math.unb.ca

Rolf Turner wrote:
> Paul E. Johnson wrote:
> 
> 
>>I handed out some results from glm() and the students ask "how many
>>observations were dropped due to missing values"?
>>
>>How would I know? 
...
>  It
> would be ***VERY*** easy to write a wrapper for glm() to include
> the number of dropped observations if you want to include that
> information in a ``printout''.
To wit

naPrint <- function(model,...) {
     if(!inherits(model,"lm") || !inherits(model,"glm"))
         stop("no support for class ",class(model),"\n")
     if(!is.null(model$na.action)) {
         action <-
paste("na.action:",attr(model$na.action,"class"))
         rows <- as.numeric(model$na.action)
         list(action=action,rows=rows)
     } else { #no missing values
         list(action=options()$na.action, rows=NULL)
     }
}

Building this as a generic function and its methods is left as an 
exercise ;)  So is a a prettier print method, and possibly a method for 
latex() or xtable().

Cheers

Jason
-- 
Indigo Industrial Controls Ltd.
http://www.indigoindustrial.co.nz
64-21-343-545
jasont at indigoindustrial.co.nz

Whoops.  That should be a "&&" where I put a "||".

naPrint <- function(model,...) {
     if(!inherits(model,"lm") &&
!inherits(model,"glm"))
         stop("no support for class ",class(model),"\n")
     if(!is.null(model$na.action)) {
         action <-
paste("na.action:",attr(model$na.action,"class"))
         rows <- as.numeric(model$na.action)
         list(action=action,rows=rows)
     } else { #no missing values
         list(action=options()$na.action, rows=NULL)
     }
}


-- 
Indigo Industrial Controls Ltd.
http://www.indigoindustrial.co.nz
64-21-343-545
jasont at indigoindustrial.co.nz