Re: [R] problems with integrate ... arguments

2009-04-10 Thread Duncan Murdoch
On 09/04/2009 8:07 AM, Richard Morey wrote: Hi everyone, I saw this problem dealt with here: http://markmail.org/search/list:r-project?q=integrate#query:list%3Ar-project%20integrate+page:1+mid:qczmyzr676pgmaaw+state:results but no one answered that request that I can tell. I'm having the same p

Re: [R] problems with integrate ... arguments

2009-04-09 Thread Duncan Murdoch
On 09/04/2009 10:19 AM, Ravi Varadhan wrote: Richard, Your function f() is already vectorized, and it works well (see below). Therefore, you don't need to create f2(). He did say that this was a simplified example. The problem appears to be in Vectorize, which uses match.call() to get the

Re: [R] problems with integrate ... arguments

2009-04-09 Thread Ravi Varadhan
Richard, I didn't read the last part of your email that you "do" want to Vectorize your function. Sorry about that. Here is a solution using "sapply" to vectorize your function. f <- function(x, const) 2^x + const f2 <- function(x, ...) sapply(x, function(x) f(x, ...) ) > integrate(f2, 0.5,

Re: [R] problems with integrate ... arguments

2009-04-09 Thread Ravi Varadhan
Richard, Your function f() is already vectorized, and it works well (see below). Therefore, you don't need to create f2(). f = function(x, const) 2^x + const testval = 2 > integrate(f, 0.5, 1, const=testval) 1.845111 with absolute error < 2.0e-14 Ravi. __

Re: [R] problems with integrate ... arguments

2009-04-09 Thread Duncan Murdoch
On 09/04/2009 8:07 AM, Richard Morey wrote: Hi everyone, I saw this problem dealt with here: http://markmail.org/search/list:r-project?q=integrate#query:list%3Ar-project%20integrate+page:1+mid:qczmyzr676pgmaaw+state:results but no one answered that request that I can tell. I'm having the same p