Make RE-QQ-Plot work for more complicated RE-structures

[strengejacke]

library(performance)
library(lme4)
data("sleepstudy")

m <- lmer(Reaction ~ Days + (1|Subject) + (0+Days|Subject), data = sleepstudy)
check_model(m)

[strengejacke]

2nd example:

m <- lmer(Reaction ~ Days + (1|Subject) + (1+Days|Subject), data = sleepstudy)

[mattansb]

Seems to work. Can we close?

library(performance)
library(lme4)
#> Loading required package: Matrix
data("sleepstudy")

m <- lmer(Reaction ~ Days + (1|Subject) + (0+Days|Subject), data = sleepstudy)
check_model(m)
#> Not enough model terms in the conditional part of the model to check for multicollinearity.
#> Could not compute standard errors from random effects for diagnostic plot.
#> Loading required namespace: qqplotr
#> `geom_smooth()` using formula 'y ~ x'
#> `geom_smooth()` using formula 'y ~ x'
#> `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
#> Warning: Removed 172 rows containing missing values (geom_text_repel).

m <- lmer(Reaction ~ Days + (1|Subject) + (1+Days|Subject), data = sleepstudy)
#> Warning in checkConv(attr(opt, "derivs"), opt$par, ctrl = control$checkConv, : Model is nearly unidentifiable: large eigenvalue ratio
#>  - Rescale variables?
check_model(m)
#> Not enough model terms in the conditional part of the model to check for multicollinearity.
#> Could not compute standard errors from random effects for diagnostic plot.
#> `geom_smooth()` using formula 'y ~ x'
#> `geom_smooth()` using formula 'y ~ x'
#> `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
#> Warning: Removed 173 rows containing missing values (geom_text_repel).

^{Created on 2021-01-13 by the reprex package (v0.3.0)}

[strengejacke]

Not sure if if it computes the correct QQ plots? Must look into the code.

[IndrajeetPatil]

Can we close?

What Mattan said 😅