On 10Oct 2020, at 4:30, Bingsong Zhang <bz117 at georgetown.edu> wrote:
In page 15 of glmmTMB manual, I see something about variance-covariance
matrix of random effect, possible options for struc() are diag(), ar1(),
cs() and so on. I just curious how to use this option?
On Thu, Oct 8, 2020 at 7:39 PM Ben Bolker <bbolker at gmail.com> wrote:
Can you please be a little more specific?
The vignette Mollie referred to is a good start.
However, there is no literal struc() option for structured covariance
matrices: the current list of options is {diag = 0, us, cs, ar1, ou,
exp, gau, mat, toep}. Or did you mean struc() as a place-holder for
structured covariance matrices in general?
On 10/8/20 3:41 PM, Bingsong Zhang wrote:
Hi all,
Did anyone use the struc(term | group) in glmmTMB before? Are there any
materials or examples I can read about it?
Best,
Bill
[[alternative HTML version deleted]]