Re: [R] Integrate inside function

I don't believe you did anything wrong here, just that integrate() itself isn't (can't?) be vectorized. E.g., x <- integrate(function(x) x^2, 0, 2) y <- integrate(function(x) x^2, 0, c(2,3)) identical(x, y) # FALSE -- so there's a difference. # Take a look str(x) str(y) # It's just in the call

Re: [R] Integrate inside function

Dear David and Micheal, thanks for your suggestion. Vectorize does what I need. David you suggest me that I didn't built the function in a manner that would vectorize. Could you please explain me what's wrong? Thanks, Mauro > > On Mar 15, 2012, at 6:08 AM, Mauro Rossi wrote: > >> Dear R users,

Re: [R] Integrate inside function

On Mar 15, 2012, at 6:08 AM, Mauro Rossi wrote: Dear R users, first I take this opportunity to greet all the R community for your continuous efforts. I wrote a function to calculate the pdf and cdf of a custom distribution (mixed gamma model). The function is the following: pmixedgamma

Re: [R] Integrate inside function

I believe the difficulty is that the integrate function isn't vectorized: add this line and you should be good pmixedgamma3 <- Vectorize(pmixedgamma3) Read ? Vectorize for details if you only need to vectorize certain arguments. Michael On Thu, Mar 15, 2012 at 6:08 AM, Mauro Rossi wrote: > Dea

[R] Integrate inside function

Dear R users, first I take this opportunity to greet all the R community for your continuous efforts. I wrote a function to calculate the pdf and cdf of a custom distribution (mixed gamma model). The function is the following: pmixedgamma3 <- function(y, shape1, rate1, shape2, rate2, prev)