[R-sig-ME] Issue with boundary (singular) fit: see ?isSingular

[Ben Bolker]

  From whom? Is piecewiseSEM really the only package you're using? It
disconcerts me that I can't locate the error message in any source
code I've found so far.

On Mon, Oct 4, 2021 at 3:04 PM Sasha Vasconcelos
<sasha.m.vasconcelos using gmail.com> wrote:
>
> Hi again,
>
> Just an update. I received this reply about the strange warning (Check model convergence: log-likelihood estimates lead to negative Chi-squared!)
>
> Yes, the convergence issues will lead to non-observable Chi-squared. If you remove those random components with variance close to 0, it should help.
>
>
>
> On Mon, 4 Oct 2021 at 15:23, Sasha Vasconcelos <sasha.m.vasconcelos using gmail.com> wrote:
>>
>> If there are only two years, it's not surprising that you'll get
>> estimates of zero variance for (1|Year).  I would probably make Year a
>> fixed effect.
>> I also tried that, leaving only Point as a random effect. But I still get the singularity warning. Could it be that the sample size is simply too small to handle any sort of random structure..?
>>
>>
>>    I can't find this warning message anywhere, even in the development
>> branch of piecewiseSEM:
>>
>> https://github.com/jslefche/piecewiseSEM/search?q=convergence
>>
>> ??
>>
>>  I also haven't been able to find anything about that warning message anywhere, so I've posted this same question to
>>
>> jslefche/piecewiseSEM on github and am hoping for an answer soon.
>>
>>
>>
>> On Mon, 4 Oct 2021 at 14:16, Ben Bolker <bbolker using gmail.com> wrote:
>>>
>>>
>>>
>>> On 10/4/21 10:05 AM, Sasha Vasconcelos wrote:
>>> > Hi,
>>> >
>>> > I'm running a piecewise SEM with 3 component models:
>>> >
>>> > lmer(response variable1 ~ predictors + (1|Point) + (1|Year), input_table)
>>> >
>>> > glmer(response variable2 ~ predictors + (1| Point) + (1|Year), family =
>>> > "binomial", input_table)
>>> >
>>> > glmer(response variable3 ~ predictors + (1| Point) + (1|Year), family =
>>> > "binomial", input_table)
>>> >
>>> > Because sampling involved visiting 18 points in spring of 2018 and again in
>>> > spring of 2019, I specified samping point and year as random effects.
>>>
>>>    If there are only two years, it's not surprising that you'll get
>>> estimates of zero variance for (1|Year).  I would probably make Year a
>>> fixed effect.
>>>
>>> >
>>> > When I run the model, this warning message appears:
>>> > Check model convergence: log-likelihood estimates lead to negative
>>> > Chi-squared!
>>>
>>>    I can't find this warning message anywhere, even in the development
>>> branch of piecewiseSEM:
>>>
>>> https://github.com/jslefche/piecewiseSEM/search?q=convergence
>>>
>>> ??
>>>
>>> >
>>> > This message also appears:
>>> > boundary (singular) fit: see ?isSingular
>>> >
>>> >  From what I've read about the second message, it could be due to random
>>> > effect variance estimates of zero. I checked and this happens in the 1st
>>> > and 3rd component models. In the 1st model "Point" has zero variance, and
>>> > in the 3rd model "Year" has zero variance.
>>> >
>>> > My first question is (and I apologize in advance if this is silly to ask)
>>> > whether this means that there's not really an effect coming from Point in
>>> > component model 1 and from Year in component model 2? If so, would it be
>>> > possible to remove those random effects to end up with:
>>> >
>>> > lmer(Response variable1 ~ Predictors + (1|Year), input_table)
>>> >
>>> > glmer(Response variable2 ~Predictors + (1| Point) + (1|Year), family =
>>> > "binomial", input_table)
>>> >
>>> > glmer(Response variable3 ~ Predictors + (1| Point), family = "binomial",
>>> > input_table)
>>>
>>>    Seems reasonable.
>>> >
>>> > My second question is whether the warning "Check model convergence:
>>> > log-likelihood estimates lead to negative Chi-squared!" is related to these
>>> > singularity issues?
>>> >
>>> > Oh and I am using the development version of the piecewise SEM package
>>> > installed using devtools. This is because this version provides additional
>>> > standardized coefficients for GLMM.
>>> >
>>> >
>>> > Thanks!
>>> >
>>> >
>>>
>>> --
>>> Dr. Benjamin Bolker
>>> Professor, Mathematics & Statistics and Biology, McMaster University
>>> Director, School of Computational Science and Engineering
>>> Graduate chair, Mathematics & Statistics
>>>
>>> _______________________________________________
>>> R-sig-mixed-models using r-project.org mailing list
>>> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
>>
>>
>>
>> --
>> Sasha Vasconcelos
>>
>> PhD student
>> CIBIO/InBIO, Research Center in Biodiversity and Genetic Resources, Associate Laboratory
>> Instituto Superior de Agronomia
>> Tapada da Ajuda
>> 1349-017 Lisbon, Portugal
>>
>> ResearchGate
>> ResearcherID
>>
>
>
> --
> Sasha Vasconcelos
>
> PhD student
> CIBIO/InBIO, Research Center in Biodiversity and Genetic Resources, Associate Laboratory
> Instituto Superior de Agronomia
> Tapada da Ajuda
> 1349-017 Lisbon, Portugal
>
> ResearchGate
> ResearcherID
>