R help - Sep 2008 - Unexpected returned value from a function

Hi R-Users,
 
I wrote a simple function to change values of a matrix or vector to NA
based on the element value being -9999 or -999999. I don't understand
why the function returns a unit vector (NA) instead of setting all
values in the vector which have -9999 or -999999 to NA. When I apply the
function in line, it appears to work correctly?

Can someone enlighten me what I am doing wrong?

Thanks in advance.

David


Here's my example:

 
ConvertMissingToNA <- function (values) {
  return ( values[ values == -9999 | values == -999999] <- NA )
}

d <- floor(runif(10, 1, 100))
pos <- floor (runif(5, 1, 10))
d[pos] <- -9999
pos <- floor (runif(2, 1, 10))
d[pos] <- -999999 
print (d)

# now apply function 
e <- ConvertMissingToNA (d)  # will return NA as a unit vector
print (e)

# conduct function in-line 
d[ d == -9999 | d == -999999] <- NA # correctly converts values to NA
print (d)

try this -- you have to return the entire vector:

ConvertMissingToNA <- function (values) {
 values[ values == -9999 | values == -999999] <- NA
 values
}

d <- floor(runif(10, 1, 100))
pos <- floor (runif(5, 1, 10))
d[pos] <- -9999
pos <- floor (runif(2, 1, 10))
d[pos] <- -999999
print (d)

# now apply function
e <- ConvertMissingToNA (d)  # will return NA as a unit vector
print (e)

# conduct function in-line
d[ d == -9999 | d == -999999] <- NA # correctly converts values to NA
print (d)

On Tue, Sep 16, 2008 at 6:19 PM, Hutchinson,David [PYR]
<David.Hutchinson at ec.gc.ca> wrote:
> Hi R-Users,
>
> I wrote a simple function to change values of a matrix or vector to NA
> based on the element value being -9999 or -999999. I don't understand
> why the function returns a unit vector (NA) instead of setting all
> values in the vector which have -9999 or -999999 to NA. When I apply the
> function in line, it appears to work correctly?
>
> Can someone enlighten me what I am doing wrong?
>
> Thanks in advance.
>
> David
>
>
> Here's my example:
>
>
> ConvertMissingToNA <- function (values) {
>  return ( values[ values == -9999 | values == -999999] <- NA )
> }
>
> d <- floor(runif(10, 1, 100))
> pos <- floor (runif(5, 1, 10))
> d[pos] <- -9999
> pos <- floor (runif(2, 1, 10))
> d[pos] <- -999999
> print (d)
>
> # now apply function
> e <- ConvertMissingToNA (d)  # will return NA as a unit vector
> print (e)
>
> # conduct function in-line
> d[ d == -9999 | d == -999999] <- NA # correctly converts values to NA
> print (d)
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>


-- 
Jim Holtman
Cincinnati, OH
+1 513 646 9390

What is the problem that you are trying to solve?

Quoting "Hutchinson,David [PYR]" <David.Hutchinson at ec.gc.ca>:
> I wrote a simple function to change values of a matrix or vector to NA
> based on the element value being -9999 or -999999. I don't understand
> why the function returns a unit vector (NA) instead of setting all
> values in the vector which have -9999 or -999999 to NA. When I apply the
> function in line, it appears to work correctly?
>
> Can someone enlighten me what I am doing wrong?
>   return ( values[ values == -9999 | values == -999999] <- NA )
the assignment itself evaluates to the assigned value, in this case NA;
that's what your function returns. Check this:
> foo <- (a <- 3)
> a
[1] 3
> foo
[1] 3
> foo <- (a <- NA)
> a
[1] NA
> foo
[1] NA

this way your function should work as intended:

ConvertMissingToNA <- function(values)
{
     values[ values == -9999 | values == -999999] <- NA
     return values
}

Peter

What you want is

ConvertMissingToNA <- function (values) {
   values[ values == -9999 | values == -999999] <- NA
   return( values )
}

To see why your version doesn't do what you wanted, maybe it helps to
consider the following?

x <- 1:10
y <- (x[3:6] <- 99)
y ## 99

(It's perhaps not entirely obvious that the value of y should be 99
and not c(99,99,99,99), but anyway, neither correspond to what you
wanted.)

Dan


On Tue, Sep 16, 2008 at 03:19:53PM -0700, Hutchinson,David [PYR]
wrote:
> Hi R-Users,
>  
> I wrote a simple function to change values of a matrix or vector to NA
> based on the element value being -9999 or -999999. I don't understand
> why the function returns a unit vector (NA) instead of setting all
> values in the vector which have -9999 or -999999 to NA. When I apply the
> function in line, it appears to work correctly?
> 
> Can someone enlighten me what I am doing wrong?
> 
> Thanks in advance.
> 
> David
> 
> 
> Here's my example:
> 
>  
> ConvertMissingToNA <- function (values) {
>   return ( values[ values == -9999 | values == -999999] <- NA )
> }
> 
> d <- floor(runif(10, 1, 100))
> pos <- floor (runif(5, 1, 10))
> d[pos] <- -9999
> pos <- floor (runif(2, 1, 10))
> d[pos] <- -999999 
> print (d)
> 
> # now apply function 
> e <- ConvertMissingToNA (d)  # will return NA as a unit vector
> print (e)
> 
> # conduct function in-line 
> d[ d == -9999 | d == -999999] <- NA # correctly converts values to NA
> print (d)
> 
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
-- 
http://www.stats.ox.ac.uk/~davison

Sorry, there was a stupid cut & paste mistake (missing parentheses in
return statement...)

ConvertMissingToNA <- function(values)
{
     values[values == -9999 | values == -999999] <- NA
     return(values)
}


Peter