[ADMB Users] adromb() ; numerical integration for RE models

[dave fournier]

On 12-11-13 09:16 AM, H. Skaug wrote:


There is something else wrong with that code. It is horrible old C style 
code
with unnecessary optimizations that make it hard to port.
> Hi,
>
> Last week it was pointed out that adromb() does not work for models
> with random effects. Dave has kindly written code to fixed this (see
> attahced reredo.zip) but there must be something wrong
> in my settup, because it does not run properly for me (but it does
> for Dave), hence I do not want to submit to the code to the trunk.
>
> Dave is doing a big service to the community by providing code
> like this very rapidly. I thus feel sorry that I am not able
> to submit it. If somebody else are able to do it would be great.
>
> Hans