Fixed effects in nlmer

Hi,

I would like to use nlmer to fit logistic growth curves to data where a number of replicate growth series are available for several clonal lines. Fitting the random effect structure is straightforward:

fm1 <- lme4:::nlmer(y ~ SSlogis(x, Asym, xmid, scal) ~  (Asym+xmid+scal | line) + (Asym+xmid+scal | rep), ?.)

allowing all 3 growth parameters to vary across lines and across replicates within lines.  However, I?ve had no success adding fixed effects to the model formula, and the examples on the help page do not have fixed effect predictors. For example, if lines could be divided into two groups (A and B) how would I allow Asym to differ between these two groups?

Thanks for any help,

Jarrod








The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336. Is e buidheann carthannais a th? ann an Oilthigh Dh?n ?ideann, cl?raichte an Alba, ?ireamh cl?raidh SC005336.

Hi.

How about adding it via an interaction effect?
Something like: y ~ SSlogis(x, Asym, Asym:lines_cathegorical, xmid, scal)
The rest should be more or less the same.

Hope that it make sense.


Cheers,

Marko


 ?On 17. 06. 2024. 15:05, Jarrod Hadfield wrote:

Hi,

I would like to use nlmer to fit logistic growth curves to data where 
a number of replicate growth series are available for several clonal 
lines. Fitting the random effect structure is straightforward:

fm1 <- lme4:::nlmer(y ~ SSlogis(x, Asym, xmid, scal) ~  
(Asym+xmid+scal | line) + (Asym+xmid+scal | rep), ?.)

allowing all 3 growth parameters to vary across lines and across 
replicates within lines.? However, I?ve had no success adding fixed 
effects to the model formula, and the examples on the help page do not 
have fixed effect predictors. For example, if lines could be divided 
into two groups (A and B) how would I allow Asym to differ between 
these two groups?

Thanks for any help,

Jarrod








The University of Edinburgh is a charitable body, registered in 
Scotland, with registration number SC005336. Is e buidheann 
carthannais a th? ann an Oilthigh Dh?n ?ideann, cl?raichte an Alba, 
?ireamh cl?raidh SC005336.

 ?? Nice idea, but I don't see how that would work.? The terms in a 
nonlinear mixed model aren't automatically expanded using linear model 
formulas ...

On 2024-06-17 9:17 a.m., marKo via R-sig-mixed-models wrote:

Hi.

How about adding it via an interaction effect?
Something like: y ~ SSlogis(x, Asym, Asym:lines_cathegorical, xmid, scal)
The rest should be more or less the same.

Hope that it make sense.


Cheers,

Marko


??On 17. 06. 2024. 15:05, Jarrod Hadfield wrote:

Hi,

I would like to use nlmer to fit logistic growth curves to data where 
a number of replicate growth series are available for several clonal 
lines. Fitting the random effect structure is straightforward:

fm1 <- lme4:::nlmer(y ~ SSlogis(x, Asym, xmid, scal) ~ 
(Asym+xmid+scal | line) + (Asym+xmid+scal | rep), ?.)

allowing all 3 growth parameters to vary across lines and across 
replicates within lines.? However, I?ve had no success adding fixed 
effects to the model formula, and the examples on the help page do 
not have fixed effect predictors. For example, if lines could be 
divided into two groups (A and B) how would I allow Asym to differ 
between these two groups?

Thanks for any help,

Jarrod








The University of Edinburgh is a charitable body, registered in 
Scotland, with registration number SC005336. Is e buidheann 
carthannais a th? ann an Oilthigh Dh?n ?ideann, cl?raichte an Alba, 
?ireamh cl?raidh SC005336.