# Re: [R] Vectorization Problem

From: David Winsemius <dwinsemius_at_comcast.net>
Date: Sat, 22 Mar 2008 16:59:57 +0000 (UTC)

"Sergey Goriatchev" <sergeyg_at_gmail.com> wrote in news:7cb007bd0803220542h66cfcd9awfd0a7a054d0fdaf8_at_mail.gmail.com:

> I have the code for the bivariate Gaussian copula. It is written
> with for-loops, it works, but I wonder if there is a way to
> vectorize the function.
> I don't see how outer() can be used in this case, but maybe one can
> use mapply() or Vectorize() in some way? Could anyone help me,
>
> ## Density of Gauss Copula
snipped your code that you didn't like

When Yan built his copula package, he called the dmvnorm function from Leisch's mvtnorm package:

dnormalCopula <- function(copula, u) {
dim <- copula_at_dimension
sigma <- getSigma(copula)
if (is.vector(u)) u <- matrix(u, ncol = dim)   x <- qnorm(u)
val <- dmvnorm(x, sigma = sigma) / apply(x, 1, function(v) prod(dnorm (v)))
val[apply(u, 1, function(v) any(v <= 0))] <- 0   val[apply(u, 1, function(v) any(v >= 1))] <- 0   val
}

If the mvtnorm package is installed, one looks at the dmvnorm function simply by typing:

dmvnorm

I did not see any for-loops. After error checking, Leisch's code is:

distval <- mahalanobis(x, center = mean, cov = sigma) logdet <- sum(log(eigen(sigma, symmetric = TRUE,
```                         only.values = TRUE)\$values))
```
logretval <- -(ncol(x) * log(2 * pi) + logdet + distval)/2 if (log)

return(logretval)
exp(logretval)

```--
David Winsemius

______________________________________________
R-help_at_r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help