Skip to content

glmer function in lme4 - false convergence warning message, (Austin, Peter)

1 message · Highland Statistics Ltd

#
------------------------------

Message: 6
Date: Sat, 22 Feb 2014 15:51:55 +0000
From: "Austin, Peter" <peter.austin at ices.on.ca>
To: "r-sig-mixed-models at r-project.org"
	<r-sig-mixed-models at r-project.org>
Cc: "Austin, Peter" <peter.austin at ices.on.ca>
Subject: [R-sig-ME] glmer function in lme4 - false convergence warning
	message
Message-ID:
	<9ED811556BF51A4C8E7FD77C5133339B45483532 at ICESMBX2.ices.on.ca>
Content-Type: text/plain; charset="iso-8859-1"

I am writing with a question about a warning message that I received when I used the glmer function to fit a complementary log-log binary regression model that incorporated two different sets of random effects. When running the model, I received the following warning message:

Warning message: In mer_finalize(ans) : false convergence (8).

However, output was produced, including regression coefficients, their associated standard errors, z values, and p-values. From the output, everything looked normal. How should I interpret the warning message, and what actions, if any, should I take?

Thanks very much,

Peter
----



Standardize your continous covariates...most of the times that helps.

Alain Zuur