Warning "In log(tmp) : NaNs produced" in MSEM

[Julian Scharbert]

Hi all,

I have fitted multiple multilevel SEMs with lavaan in multiple datasets, respectively, which all converge without problems and provide reasonable outputs. One model produces a warning message, though, which I do not know how to address. The same model converges in the other datasets and in the same dataset with other predictors, but not in this specific case. I have encountered a comparable warning message in this thread and the author of the post told me that the problem was caused by him using an outdated version of lavaan. I have just installed the newest version of lavaan via 

install.packages("lavaan", repos = "https://www.da.ugent.be/", type = "source")

and the newest r-version, which did not fix the problem, though. 

The model I am trying to fit looks like this:

sem(formula, data=data, cluster="participants")

formula = 

"level:1

l1_outcome~b*l1_predictor

level:2

l1_outcome~c*l2_moderator

l1_predictor~a*l2_moderator

ab:= a*b

total:= ab+c"

The warning message I receive is this (multiple times):

"In log(tmp) : NaNs produced"

I understand that this problem is likely caused by the log() function being applied to a negative value, but I do not understand lavaan well enough to understand why this is the case for the tmp-value. 

Does someone have an idea how to address this problem? The model output looks reasonable to me as well and is comparable to the other predictors / the same predictor with other datasets. So if the warning message can be safely ignored, that would also work for me. 

Thank you in advance! 

[Yves Rosseel]

Hello Julian,

I would like to investigate this further. Would you be able to email me
your dataset (or a snippet of the datset) and a small R script, so I can
reproduce this?

Please send it to Yves dot Rosseel at UGent dot be

Yves.

On 9/1/22 11:42, 'Julian Scharbert' via lavaan wrote:
> Hi all,
>
> I have fitted multiple multilevel SEMs with lavaan in multiple datasets,
> respectively, which all converge without problems and provide reasonable
> outputs. One model produces a warning message, though, which I do not
> know how to address. The same model converges in the other datasets and
> in the same dataset with other predictors, but not in this specific
> case. I have encountered a comparable warning message in this

> <https://groups.google.com/g/lavaan/c/L0my1YG89K8/m/s6cV7fj4CAAJ>thread

> and the author of the post told me that the problem was caused by him
> using an outdated version of lavaan. I have just installed the newest
> version of lavaan via

> /i//nstall.packages("lavaan", repos = "https://www.da.ugent.be/
> <https://www.da.ugent.be/>", type = "source")/
> //and the newest r-version, which did not fix the problem, though.

>
> The model I am trying to fit looks like this:

> /sem(formula, data=data, cluster="participants")
> /
> /formula = /
> /"level:1/
> /l1_outcome~b*l1_predictor/
> /level:2/
> /l1_outcome~c*l2_moderator/
> /l1_predictor~a*l2_moderator/
> /ab:= a*b/
> /total:= ab+c"/

>
> The warning message I receive is this (multiple times):

> /"In log(tmp) : NaNs produced"/

>
> I understand that this problem is likely caused by the log() function
> being applied to a negative value, but I do not understand lavaan well
> enough to understand why this is the case for the tmp-value.
>
> Does someone have an idea how to address this problem? The model output
> looks reasonable to me as well and is comparable to the other predictors
> / the same predictor with other datasets. So if the warning message can
> be safely ignored, that would also work for me.
>
> Thank you in advance!
>

> --
> You received this message because you are subscribed to the Google
> Groups "lavaan" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to lavaan+un...@googlegroups.com
> <mailto:lavaan+un...@googlegroups.com>.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/lavaan/9ad9d58a-2561-4294-ae81-220820ca9bebn%40googlegroups.com
> <https://groups.google.com/d/msgid/lavaan/9ad9d58a-2561-4294-ae81-220820ca9bebn%40googlegroups.com?utm_medium=email&utm_source=footer>.

[Julian Scharbert]

Hi all, 

Yves helped me figure out the problem. I had a dev version of lavaan installed (0.6-11.1676). After switching to the current CRAN version (0.6-12), the model did not produce the warning message anymore. Maybe this helps someone with a related problem.

Julian