Fetch tidy names in grouped mixed-model

[alapo]

This issue is related to Issue 21.

In the example below I want to loop through Species which I can do using grouped_tidy or grouped_glance
My goal is to obtain the p-values that I can obtain using parameters::model_parameters()

# Grouped analysis (does not provide p-value)
tmp <- broomExtra::grouped_glance(
  data = iris,
  grouping.vars = c(Species),
  formula = Sepal.Length ~ Petal.Length + Petal.Width + Petal.Length*Petal.Width + (1|Sepal.Width),
  ..f = lme4::lmer,
  control = lme4::lmerControl(optimizer = "bobyqa")
)

# Ungrouped analyses (does not loop)
fm1 <- lmer(Sepal.Length ~ Petal.Length + Petal.Width + Petal.Length*Petal.Width + (1|Sepal.Width), data = iris %>% filter(Species == unique(Species)[1]))
   parameters::model_parameters(fm1) %>%
   broomExtra::easystats_to_tidy_names()

I tried using the code supplied on the GitHubIO but was unable to fetch p-values. I tried 2 versions of tidy.args

# Version 1 of tidy_args
tmp <- broomExtra::grouped_tidy(
  data = iris,
  grouping.vars = c(Species),
  formula = Sepal.Length ~ Petal.Length + Petal.Width + Petal.Length*Petal.Width + (1|Sepal.Width),
  ..f = lme4::lmer,
  control = lme4::lmerControl(optimizer = "bobyqa"),
  tidy.args = list(quick = TRUE)
)

# Version 2 of tidy_args
tmp <- broomExtra::grouped_tidy(
  data = iris,
  grouping.vars = c(Species),
  formula = Sepal.Length ~ Petal.Length + Petal.Width + Petal.Length*Petal.Width + (1|Sepal.Width),
  ..f = lme4::lmer,
  control = lme4::lmerControl(optimizer = "bobyqa"),
  tidy.args = list(conf.int = TRUE, conf.level = 0.99)
  )

P.S: Was going to post on GitHub but there are not any broomExtra tags permitted at the moment so it seemed best to post here to get a response. Apologies if this is a mundane question

[IndrajeetPatil]

Better to use dplyr::group_modify here

set.seed(123)
library(lme4)
library(dplyr)
options(tibble.width = Inf)

iris %>%
  group_by(Species) %>%
  group_modify(., ~ broomExtra::easystats_to_tidy_names(as.data.frame(parameters::model_parameters(
    lme4::lmer(Sepal.Length ~ Petal.Length + Petal.Width + Petal.Length * Petal.Width + (1 | Sepal.Width),
      data = .x,
      control = lme4::lmerControl(optimizer = "bobyqa")
    )
  ))))
#> Registered S3 method overwritten by 'broom.mixed':
#>   method      from 
#>   tidy.gamlss broom
#> # A tibble: 12 x 9
#> # Groups:   Species [3]
#>    Species    term                     estimate std.error conf.low conf.high
#>    <fct>      <chr>                       <dbl>     <dbl>    <dbl>     <dbl>
#>  1 setosa     (Intercept)                 3.73      0.961   1.85       5.62 
#>  2 setosa     Petal.Length                0.900     0.635  -0.344      2.14 
#>  3 setosa     Petal.Width                 3.56      3.67   -3.64      10.8  
#>  4 setosa     Petal.Length:Petal.Width   -2.34      2.35   -6.94       2.27 
#>  5 versicolor (Intercept)                 1.08      2.89   -4.58       6.74 
#>  6 versicolor Petal.Length                1.23      0.677  -0.0930     2.56 
#>  7 versicolor Petal.Width                 0.814     2.39   -3.87       5.50 
#>  8 versicolor Petal.Length:Petal.Width   -0.259     0.537  -1.31       0.793
#>  9 virginica  (Intercept)                -4.04      4.16  -12.2        4.11 
#> 10 virginica  Petal.Length                1.92      0.753   0.444      3.40 
#> 11 virginica  Petal.Width                 2.53      2.07   -1.52       6.58 
#> 12 virginica  Petal.Length:Petal.Width   -0.459     0.371  -1.19       0.269
#>    statistic df.error  p.value
#>        <dbl>    <int>    <dbl>
#>  1     3.88        44 0.000104
#>  2     1.42        44 0.156   
#>  3     0.969       44 0.332   
#>  4    -0.994       44 0.320   
#>  5     0.374       44 0.708   
#>  6     1.82        44 0.0684  
#>  7     0.340       44 0.734   
#>  8    -0.482       44 0.630   
#>  9    -0.972       44 0.331   
#> 10     2.55        44 0.0108  
#> 11     1.22        44 0.221   
#> 12    -1.23        44 0.217

^{Created on 2020-10-21 by the reprex package (v0.3.0)}

[alapo]

Such a boss! Thanks for the support.