Store several regression models in a list in R

August 29, 2023

I run several regressions (OLS/IV) on different outcome variables. I like to store all results in one list. With 2 models and 3 outcomes, the final list should have 6 elements, and be ordered like OLS1, IV1, OLS2, IV2, OLS3, IV3. Is there another way/shorter/more direct instead of creating a list of list and converting back to a normal list?

library(ivreg)
data(mtcars)

models <- c("mpg", "disp", "hp")

linear_regressions <- list()
iv_regressions <- list()

for (i in models) {
  formula1 <- paste(i, "~ cyl + qsec + carb + drat")
  formula2 <- paste(i, "~ cyl + qsec + carb | drat | wt")
  
  linear_model <- lm(formula = formula1, data = mtcars)
  iv_model <- ivreg(formula = formula2, data = mtcars)
  
  regression_list <- list(linear = linear_model, iv = iv_model)
  
  all_regressions[[i]] <- regression_list
}

flattened_regressions <- unlist(all_regressions, recursive = FALSE)

library(modelsummary)
modelsummary(flattened_regressions, output = "gt")

>Solution :

You can directly create a list where each element corresponds to an OLS and an IV regression for each outcome variable :

library(ivreg)
data(mtcars)

models <- c("mpg", "disp", "hp")

regressions <- list()

for (i in models) {
  formula1 <- paste(i, "~ cyl + qsec + carb + drat")
  formula2 <- paste(i, "~ cyl + qsec + carb | drat | wt")
  
  linear_model <- lm(formula = formula1, data = mtcars)
  iv_model <- ivreg(formula = formula2, data = mtcars)
  
  regressions[[paste0(i, ".linear")]] <- linear_model
  regressions[[paste0(i, ".iv")]] <- iv_model
}

library(modelsummary)
modelsummary(regressions, output = "gt")