Skip to content
Prev 42837 / 398506 Next

random effects with lme() -- comparison with lm()

On Thu, 2004-01-15 at 16:30, Douglas Bates wrote:
<...snip...>
Agreed.

<...snip...>
Following your suggestion, I got:
Error in intervals.lme(lme(Y ~ 1, data = simdat, random = ~1 | A)) :
        Cannot get confidence intervals on var-cov components:
Non-positive definite approximate variance-covariance

This led me to:
[1] "Non-positive definite approximate variance-covariance"

As a new feature suggestion for lme(), would it be appropriate to use
"apVar" as a warning flag in this case?

Sincerely,
Jerome Asselin