R help - Jun 2002 - using integrate on a function defined with if

Dear R People,

I getting some curious behaviour with the integrate function. Consider the
function

foo <- function(x)
  {
    if(x==0)
	4
	else
	0
	}

I get the error
> integrate(foo,0,1)
Error in integrate(foo, 0, 1) : evaluation of function gave a result of
wrong length

But now consider

foo <- function(x)
  {
    ifelse(x==0,4,0)
  }
> integrate(foo,0,1)
0 with absolute error < 0

I am guessing this may have something to do with types/classes, but I
don't see exactly what. I would be happy to be enlightened.

Yhe help page says the function must be

"f An R function taking a numeric first argument and returning a numeric
vector the same length."

But I don't see any difference in what these two return. In both cases it
is NULL.

                                  Sincerely, Faheem Mitha.

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

On Sat, 15 Jun 2002, Faheem Mitha wrote:
>
> Dear R People,
>
> I getting some curious behaviour with the integrate function. Consider the
> function
>
> foo <- function(x)
>   {
>     if(x==0)
> 	4
> 	else
> 	0
> 	}
>
> I get the error
>
> > integrate(foo,0,1)
> Error in integrate(foo, 0, 1) : evaluation of function gave a result of
> wrong length
>
> But now consider
>
> foo <- function(x)
>   {
>     ifelse(x==0,4,0)
>   }
>
> > integrate(foo,0,1)
> 0 with absolute error < 0
>
> I am guessing this may have something to do with types/classes, but I
> don't see exactly what. I would be happy to be enlightened.
>
> Yhe help page says the function must be
>
> "f An R function taking a numeric first argument and returning a
numeric
> vector the same length."
>
> But I don't see any difference in what these two return. In both cases
it
> is NULL.
No, in the first case it is vector of length 1, and in the second a vector
of the length of x.   Did you actually try it?  Please do to convince
yourself.

S functions return the value of their last statement.


-- 
Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272860 (secr)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

Hi Faheem

``integrate'' requires that the function (foo in your example) 
takes a vector as the first argument and returns a vector of the same
length. 

To illustrate this, try making a check for x being a vector in your
first example

foo <- function(x){
  if(is.vector(x)) print("x is a vector ! ")
  if(x==0) 4
  else 0
}

integrate(foo,0,1)

gives 

[1] "x is a vector ! "
Error in integrate(foo, 0, 1) : evaluation of function gave a result of
wrong length


As Prof. Ripley wrote in his reply, the problem is that your function is
not returning a vector but a number.

Cheers Ole



Faheem Mitha wrote:
> 
> Dear R People,
> 
> I getting some curious behaviour with the integrate function. Consider the
> function
> 
> foo <- function(x)
>   {
>     if(x==0)
>         4
>         else
>         0
>         }
> 
> I get the error
> 
> > integrate(foo,0,1)
> Error in integrate(foo, 0, 1) : evaluation of function gave a result of
> wrong length
> 
> But now consider
> 
> foo <- function(x)
>   {
>     ifelse(x==0,4,0)
>   }
> 
> > integrate(foo,0,1)
> 0 with absolute error < 0
> 
> I am guessing this may have something to do with types/classes, but I
> don't see exactly what. I would be happy to be enlightened.
> 
> Yhe help page says the function must be
> 
> "f An R function taking a numeric first argument and returning a
numeric
> vector the same length."
> 
> But I don't see any difference in what these two return. In both cases
it
> is NULL.
> 
>                                   Sincerely, Faheem Mitha.
> 
>
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
> 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
>
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
-- 
Ole F. Christensen
Department of Mathematics and Statistics
Fylde College, Lancaster University 
Lancaster, LA1 4YF, England
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._