Dear All, I wonder if this is a technical or an interpretation problem. I fitted an lme model including a random intercept and a random slope. The estimated correlation, from the lme, between the random intercept and the random slope is equal to 1. Does it mean that I should suppress one of the random effect from the model? Thanks a lot, Bernard --------------------------------- [[alternative HTML version deleted]]
or, it could mean you need to recenter your time variable. -----Original Message----- From: r-help-bounces@stat.math.ethz.ch on behalf of Marc Bernard Sent: Fri 11/3/2006 7:24 AM To: r-help@stat.math.ethz.ch Subject: [R] correaltion equal 1 Dear All, I wonder if this is a technical or an interpretation problem. I fitted an lme model including a random intercept and a random slope. The estimated correlation, from the lme, between the random intercept and the random slope is equal to 1. Does it mean that I should suppress one of the random effect from the model? Thanks a lot, Bernard --------------------------------- [[alternative HTML version deleted]] ______________________________________________ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. [[alternative HTML version deleted]]
On 11/3/06, Doran, Harold <HDoran at air.org> wrote:> or, it could mean you need to recenter your time variable. > > > -----Original Message----- > From: r-help-bounces at stat.math.ethz.ch on behalf of Marc Bernard > Sent: Fri 11/3/2006 7:24 AM > To: r-help at stat.math.ethz.ch > Subject: [R] correaltion equal 1 > > Dear All, > > I wonder if this is a technical or an interpretation problem. I fitted an lme model including a random intercept and a random slope. The estimated correlation, from the lme, between the random intercept and the random slope is equal to 1. Does it mean that I should suppress one of the random effect from the model?It probably indicates that the estimate of the variance-covariance matrix for the random effects is singular. If you refit the model using lmer from the lme4 package and the estimate of the variance-covariance matrix is singular, you will get a warning to that effect. As Harold mentioned, you may be able to avoid the problem by re-centering the covariate. However, this may also be a characteristic of your data.