Would you post your syntax, please. I'm confused by the this part of the warning: ... com(p)uting the inverse logit link function ...) and your statement that you are running what I think would be described as a one between one within repeated measures analysis. If you are not allowing random intercepts and have no missing data, you could run this in GLM repeated.
Gene Maguin
-----Original Message-----
From: SPSSX(r) Discussion <
[hidden email]> On Behalf Of birolus
Sent: Tuesday, May 8, 2018 9:42 AM
To:
[hidden email]
Subject: GLMM warning: At least one invalid record in the last iteration (zie message for continued warning)
The warning continues with A record is invalid if there are errors in comuting the inverse logit link function, the log-likelihood, the gradient, or the hessian matrix in the iterative process. No output will be displayed.
Execution of the command stops.
We are comparing a dichototomous independent variable with a repeted measurement of a continous variable.
Could you please advise?
Kind Regards
--
Sent from:
http://spssx-discussion.1045642.n5.nabble.com/=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD
=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD
Locked
