Skip to content
Prev 18794 / 20628 Next
Original Raw

confidence intervals with mvrnorm - upper value equal to inf

Ben Bolker 
It's hard to troubleshoot this without a reproducible example. Unless 
the answer is obvious -- which it's not, to me -- the easiest way to 
troubleshoot is to work through the steps one at a time and see where 
the infinite values first appear.  Can you create such an example either 
by posting your data or by simulating data that looks like your data?

   The posterior predictive simulation approach assumes that the 
sampling distributions of the parameters are multivariate normal, which 
is likely to be questionable in a low-information setting (which will be 
the case if you don't have very many non-zero values ...)
On 10/10/20 7:28 AM, Alessandra Bielli wrote:

Thread (8 messages)

Alessandra Bielli confidence intervals with mvrnorm - upper value equal to inf Oct 10 Ben Bolker confidence intervals with mvrnorm - upper value equal to inf Oct 11 Alessandra Bielli confidence intervals with mvrnorm - upper value equal to inf Oct 14 Ben Bolker confidence intervals with mvrnorm - upper value equal to inf Oct 14 Ben Bolker confidence intervals with mvrnorm - upper value equal to inf Oct 15 Alessandra Bielli confidence intervals with mvrnorm - upper value equal to inf Oct 15 Ben Bolker confidence intervals with mvrnorm - upper value equal to inf Oct 19 Alessandra Bielli confidence intervals with mvrnorm - upper value equal to inf Oct 19