[R-sig-ME] "Non-positive definite approximate variance-covariance" output

[Jacob Bukoski]

​Great -- that makes sense. Thank you!

Jacob​


On Feb 17, 2016 3:44 PM, "Douglas Bates" <bates at stat.wisc.edu> wrote:

> The standard errors are for the fixed-effects parameters.  The apVar
> component includes approximate variances of the estimators of (functions
> of) the variance components, which is where the positive-definiteness
> failure may be occurring..
>
> On Wed, Feb 17, 2016 at 2:28 PM Jacob Bukoski <jbukoski1 at gmail.com> wrote:
>
>> Hi all,
>>
>> I was in a discussion with my advisor today, and the issue of
>> "non-positive
>> definite approximate variance-covariance" errors in R when using the lme()
>> function arose.
>>
>> Depending on the varFunc specification that I use, I receive the
>> "non-positive definite approximate variance-covariance" output when
>> calling
>> myModel$apVar -- however, there are standard errors returned for my model
>> coefficients in the model summary.
>>
>> My understanding is that if the variance-covariance cannot be
>> approximated,
>> neither should the standard errors... how is the lme() function returning
>> standard errors without a variance-covariance approximation?
>>
>> Many thanks,
>> Jacob
>>
>> --
>> Jacob J. Bukoski
>> Master of Environmental Science Candidate, 2016
>> School of Forestry and Environmental Studies, Yale University
>> jbukoski1 at gmail.com | jacob.bukoski at yale.edu | LinkedIn
>> <
>> https://www.linkedin.com/profile/view?id=AAIAAAdWVW8BMzqU_2EGNbEkyuy8O7K1Jyhd8ps&trk=nav_responsive_tab_profile_pic
>> >
>>
>>         [[alternative HTML version deleted]]
>>
>> _______________________________________________
>> R-sig-mixed-models at r-project.org mailing list
>> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
>>
>

	[[alternative HTML version deleted]]