Model Support Request: plm::plm() models

[ddsjoberg]

This stackoverflow post is asking about support for plm::plm() models.
https://stackoverflow.com/questions/70560374

I've never used that package/model before. I'll see if I can submit a PR with support for the models tonight.

library(broom.helpers)
data("Grunfeld", package = "plm")

model <- plm::plm(inv ~ value + capital, data = Grunfeld, model = "within", index = c("firm", "year"))
model
#> 
#> Model Formula: inv ~ value + capital
#> 
#> Coefficients:
#>   value capital 
#> 0.11012 0.31007

broom::tidy(model)
#> # A tibble: 2 x 5
#>   term    estimate std.error statistic  p.value
#>   <chr>      <dbl>     <dbl>     <dbl>    <dbl>
#> 1 value      0.110    0.0119      9.29 3.92e-17
#> 2 capital    0.310    0.0174     17.9  2.22e-42

tidy_plus_plus(model)
#> Error: Tibble columns must have compatible sizes.
#> * Size 2: Existing data.
#> * Size 3: Column `variable`.
#> i Only values of size one are recycled.

^{Created on 2022-01-02 by the reprex package (v2.0.1)}

[ddsjoberg]

UPDATE: The issue root is that the plm() function does not return an intercept. I was able to add model_identify_variables.plm() that fixed the issue with the missing intercept. But I got stuck with a missing reference row. When you get a chance, can you take a look?

[larmarange]

I will look at it once back Joseph Larmarange Le lun. 3 janv. 2022 à 05:06, Daniel Sjoberg ***@***.***> a écrit :

…

UPDATE: The issue root is that the plm() function does not return an intercept. I was able to add model_identify_variables.plm() that fixed the issue with the missing intercept. But I got stuck with a missing reference row. When you get a chance, can you take a look? — Reply to this email directly, view it on GitHub <#140 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAHL5I6LJGSS4SGSWDD4JZTUUDZBBANCNFSM5LEF63JQ> . Triage notifications on the go with GitHub Mobile for iOS <https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675> or Android <https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub>. You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[larmarange]

OK, I have identified the issue. By default, plm:::model.matrix.plm() removes constant columns (i.e. the intercept) resulting in a bug when assigning variable (because the assign attribute was not updated accordingly).

model_get_model_matrix.plm <- function(model, ...) {
  stats::model.matrix(model, cstcovar.rm = "none", ...)
}

should be enough to fix the bug