*****  To join INSNA, visit http://www.insna.org  *****

Hi Zachary,

It seems that you are trying to fit models similar to those in
https://urldefense.proofpoint.com/v2/url?u=https-3A__link.springer.com_article_10.1353_dem.0.0045&d=DwIFaQ&c=sJ6xIWYx-zLMB3EPkvcnVg&r=yQQsvTNAnbvDXGM4nDrXAje4pr0qHX2qIOcCQtJ5k3w&m=hdKFP6SsA-N-KhBQHnJJaknlNxphcgwIN7PaU2SFDgQ&s=oG5dVZEew7kwU07oPkSWSCZ3nEGwfZTA3KUE-iPb5Y8&e= , correct?

I think the first question is what specifications without GWESP did
you try and how their goodness of fit look like? There are many things
that might cause non-convergence errors. Stalled iterations worry me
actually much more... If your model-fitting uses adaptive MCMC it may
be basically also a symptom of non-convergence as the algorithm might
be trying to increase the MCMC sample size to very big numbers which
might look as a "stalled iteration".

Second somewhat standard technical question is: are you using the most
recent version of the ergm package and friends?

Best,
Michal

On Thu, Jun 4, 2020 at 5:07 PM Neal, Zachary <[log in to unmask]> wrote:
>
> ***** To join INSNA, visit https://urldefense.proofpoint.com/v2/url?u=http-3A__www.insna.org&d=DwIFaQ&c=sJ6xIWYx-zLMB3EPkvcnVg&r=yQQsvTNAnbvDXGM4nDrXAje4pr0qHX2qIOcCQtJ5k3w&m=hdKFP6SsA-N-KhBQHnJJaknlNxphcgwIN7PaU2SFDgQ&s=Teh6GBwpDFUGM34ZZXBtf_w52WB6qI3vW_ssHTrRPDM&e=  *****
> Hi all,
>
> I am hoping to get some assistance or pointers on estimating a problematic ERGM using statnet::ERGM in R.
>
> There are multiple networks, but each one contains about 450 nodes, with a density of about 0.1. The parameters of interest are estimated using a few nodematch(diff=TRUE) terms, so it’s a pretty basic model.
>
> The problem is that including gwesp(decay=0, fixed=TRUE) yields either a failed convergence or stalled iteration. Alternative decay parameters do not seem to help. I suspect part of the problem may be that these networks are highly transitive (global transitivity ~ 0.6).
>
> I would welcome any suggestions on paths forward (e.g. achieving convergence with GWESP, alternatives to GWESP, need for GWESP when transitivity is known to exist and isn’t of interest, etc.)
>
> Thanks,
> Zachary
>
> –––
> Zachary Neal, PhD
> Associate Professor, Michigan State University
> Web: https://urldefense.proofpoint.com/v2/url?u=http-3A__www.zacharyneal.com&d=DwIFaQ&c=sJ6xIWYx-zLMB3EPkvcnVg&r=yQQsvTNAnbvDXGM4nDrXAje4pr0qHX2qIOcCQtJ5k3w&m=hdKFP6SsA-N-KhBQHnJJaknlNxphcgwIN7PaU2SFDgQ&s=qeDeYbrO2gYTjKF9PTbhH1C6LSvowttUq7kj3iQCLOw&e= 
> Twitter: @zpneal
> Zoom: Click here
>
> _____________________________________________________________________ SOCNET is a service of INSNA, the professional association for social network researchers (https://urldefense.proofpoint.com/v2/url?u=http-3A__www.insna.org&d=DwIFaQ&c=sJ6xIWYx-zLMB3EPkvcnVg&r=yQQsvTNAnbvDXGM4nDrXAje4pr0qHX2qIOcCQtJ5k3w&m=hdKFP6SsA-N-KhBQHnJJaknlNxphcgwIN7PaU2SFDgQ&s=Teh6GBwpDFUGM34ZZXBtf_w52WB6qI3vW_ssHTrRPDM&e= ). To unsubscribe, send an email message to [log in to unmask] containing the line UNSUBSCRIBE SOCNET in the body of the message.

_____________________________________________________________________
SOCNET is a service of INSNA, the professional association for social
network researchers (http://www.insna.org). To unsubscribe, send
an email message to [log in to unmask] containing the line
UNSUBSCRIBE SOCNET in the body of the message.