[R-sig-ME] Estimating multiple membership models with lme4

[Sijia Huang]

Hi all,
I am trying to estimate a multiple membership multiple classification model
with lme4. What I have in the model is that individuals are
cross-classified by ego nets and schools, and each individual can belong to
more than one ego nets.
I have following the instruction,
https://bbolker.github.io/mixedmodels-misc/notes/multimember.html

Below are my codes, where W is a matrix of 0s and 1s that indicates if an
individual belongs to an ego net.

lmod <- lFormula(y~(1|fake.ego.id)+(1|sch.id), data=data)
lmod$reTrms$Zt <- lmod$reTrms$Ztlist[[1]] <- Matrix(t(W))
devfun <- do.call(mkLmerDevfun, lmod)

Error in Lambdat %*% Ut :
  Cholmod error 'A and B inner dimensions must match' at file
../MatrixOps/cholmod_ssmult.c, line 82

Anyone have an idea how to fix this error? Thanks!


Best,
Sijia

	[[alternative HTML version deleted]]