How to create a list of linear models in R?

Question

I need to create a named list of linear models in R.

models=list()
for (tag in tagnames){
   expr=paste0(tag," ~ .")
   f=formula(expr)
   models[tag]=lm(f,df)
 }

This is the code that I wrote; it actually creates the list, but apparently it is a list of lists which are not callable objects (i.e., the method predict doesn't work on the elements of the list).

How can I make the element of the lists usable as actual lm objects?

Answer 1

You can create a list of your models with lapply:

models <- lapply(tagnames, function(x) lm(formula(paste0(x, " ~ .")), df))

and assign the names with

names(models) <- tagnames

Then call predict on the list element:

predict(models[["name"]])