estimating common factor scores at the level of a site

[dougwyu]

hello!

I have been running an unconstrained gllvm with a row effect, and of course, gllvm returns LV scores per row (= per sample). What i would like are LV scores on a per site (e.g. per Lake) basis. Is that possible? The ugliest way i could think of to achieve this is to take the mean of each LV per site.

doug

gllvm( y = y, family = binomial() , num.lv = lvchosen, studyDesign = Site, row.eff = ~ (1 | Lake), control.start = list(n.init = 10) )

[BertvanderVeen]

Thanks Doug.

This is currently not possible in gllvm, and I am not sure it is on the to-do list either. But @JenniNiku might be able to provide more details.

[JenniNiku]

Hi, I'm currently working on structured LVs, like this