How to predict, using plm()?

[OleksiyAnokhin]

Hi team, thanks for a great package?
I check all possible resources, but do not see how to used predict() or prediction() with plm.

Currently I am trying:

predictions <- predict(plmFit1, panel_df_test)

and get

Error in crossprod(beta, t(X)) : non-conformable arguments

[tappek]

Hi, seems like you were trying to predict using a within model. That case what not implemented - until now in the development version (commits: b81d759, 93a51bd, 4c44fef, cd48d6f) .

Be aware that the within case is somewhat special, more so with out-of-sample prediction. The help page ?predict.plm has the details and examples.

Feedback is welcome!

To install the development version from GitHub:

# install.packages("remotes") # remove '#' if pkg 'remotes' is not installed
remotes::install_github("ycroissant/plm")

[tappek]

Some reference asking for predict for plm objects/fixed effects models (partly copied from jacob-long/jtools#97 (comment)):

leeper/prediction#1
leeper/prediction#19
strengejacke/ggeffects#121
strengejacke/sjPlot#594
jacob-long/jtools#97
tidymodels/broom#73
https://stackoverflow.com/questions/31954045/r-plotting-panel-model-predictions-using-plm-pglm
https://stackoverflow.com/questions/64892622/plotting-interaction-terms-from-fixed-effects-models-plm
https://stackoverflow.com/questions/71366468/predict-on-test-data-using-plm-package-in-r-and-calculate-rmse-for-test-data
https://stackoverflow.com/questions/71894726/how-can-effects-from-fixed-effects-model-in-plm-be-plotted-in-r
https://stackoverflow.com/q/65702581/4640346
https://stackoverflow.com/questions/70150063/how-can-i-use-predict-to-a-plm-regression-model
https://stackoverflow.com/questions/14427781/how-to-deal-with-na-in-a-panel-data-regression
https://stackoverflow.com/a/73127507/4640346
https://stats.stackexchange.com/questions/137923/prediction-with-plm-method
https://stats.stackexchange.com/questions/147350/computing-the-predicted-value-from-a-panel-data-model-with-the-plm-r-package?rq=1

[tappek]

Closing this because it is solved.

[tappek]

On CRAN as plm 2.6-2 as of 2022-08-16