"R session aborted" when using gofN()

[liluqi1991]

Hello, I have been using ergm.multi in my manuscript and a reviewer has requested a diagnostic analysis. However, whenever I employ _Layer()_ to define a multilayer network, the “R Session Aborted” error occurs after executing gofN(). I have noticed that in Krivitsky’s study (2020, https://doi.org/10.1007/s11336-020-09720-7), the goodness-of-fit assessment for the Lazega example was successfully conducted, and it is likely that the model was specified using Layer(). I would greatly appreciate any assistance in solving this issue.

library(ergm.multi)
data(florentine)

# Work:
flo <- Networks(m = flomarriage, b = flobusiness)
fit <- ergm(flo~N(~edges))
gof<-gofN(fit)

# Not work (R session aborted):
flo <- Layer(m = flomarriage, b = flobusiness)
fit <- ergm(flo ~ L(~edges, ~m)+L(~edges, ~b))
gof<-gofN(fit)

[krivit]

gofN() is for samples of networks. It doesn't work for multilayer models, nor should it, since layers are generally not independent. Diagnostics for multilayer models are on the roadmap, but it'll be a while yet.

[krivit]

To add to this, the diagnostics for the paper were done using bespoke code, and I haven't had a chance to make it general.

[krivit]

And thank you, @liluqi1991, for flagging this issue.

[liluqi1991]

And thank you, @liluqi1991, for flagging this issue.

Thank you very much for your reply.