issue with new mixedup or new mgcv

[BenoitLondon]

Hi, first thank you for that great package! Unfortunately, it does not work for my use case anymore :( Hope you can help :D Here's a reprex:

packageVersion("mixedup")
# [1] ‘0.4.0’
packageVersion("mgcv")
# [1] ‘1.8.41’
mtcarsf <- mtcars
mtcarsf$cyl <- factor(mtcarsf$cyl)
mod <- mgcv::bam(mpg ~ s(cyl, bs = "re", by = I(log(wt))), data = mtcarsf)
coefs <- mixedup::extract_random_effects(mod)
# Error in `[.data.frame`(model$model, , re_names[i]) : 
# undefined columns selected

digging into the code vn does not look to contain a valid column anymore when grabbed here

$vn
[1] "eraI(log(expected_dog_goals))"

if no by term in the spline, it works fine

> mod <- mgcv::bam(mpg ~ s(cyl, bs = "re"), data = mtcarsf)
> coefs <- mixedup::extract_random_effects(mod)
> coefs
# A tibble: 3 × 7
  group_var effect    group value    se lower_2.5 upper_97.5
  <chr>     <chr>     <chr> <dbl> <dbl>     <dbl>      <dbl>
1 cyl       Intercept 4      6.00  3.42     -0.71      12.7 
2 cyl       Intercept 6     -0.72  3.45     -7.48       6.04
3 cyl       Intercept 8     -5.28  3.41    -12.0        1.41

[m-clark]

I don't think it has to do with the latest mixedup, it's just that I don't have functionality for 'by' stuff (except maybe by accident). But I think it's also that you can do this more clearly as a random slope of log_wt across cyl approach

From your primary model:

> coef(mod)
        (Intercept) s(cyl):I(log(wt)).1 s(cyl):I(log(wt)).2 s(cyl):I(log(wt)).3 
          33.933241           -9.382861          -12.449715          -13.648564 

Differently expressed:

> mod <- mgcv::bam(mpg ~ s(log_wt, cyl, bs = "re"), data = mtcarsf)
> coefs <- mixedup::extract_random_effects(mod)
> coefs
# A tibble: 3 × 7
  group_var effect group  value    se lower_2.5 upper_97.5
  <chr>     <chr>  <chr>  <dbl> <dbl>     <dbl>      <dbl>
1 cyl       log_wt 4      -9.38  2.66     -14.6      -4.17
2 cyl       log_wt 6     -12.4   2.08     -16.5      -8.37
3 cyl       log_wt 8     -13.6   1.64     -16.9     -10.4 

Same result. 😄 I also think that model would be more clear to you in the future as well.

Other notes:

• I'd also recommend adding a separate 're' for your factor variable so that you have the random intercept and slope as separate effects- i.e. + s(cyl, bs = 're'). You can compare a model with and without the random slope and possibly simplify the model if need be, but mainly that would be more consistent with how mixed models with random slopes are usually expressed.
• I would avoid passing I() or log(), as mgcv and/or mixedup (and many other packages) will likely have naming issues at some point. Just create the feature separately as log_wt to avoid potential problems.

Hope that helps, but let me know if you have other issues.

[m-clark]

Going ahead and closing but feel free to reopen if needed.

[BenoitLondon]

thanks that all makes sense.