4 Why Regression?

We’ve discussed the “good” vs. “bad” pathways in the previous lecture. What we’d want to do is close off the “bad” pathways as much as possible. One way to do this is by using regression.

In fact, we ran several regression models towards the end of the last lecture. And we were happy that by controlling for income, we were able to trace the effect of college education on health as defined by the simulated DGP. By accounting for the necessary variables, regression may allow us to estimate causal effects. However, there are strong limitations to this approach, something which will discuss soon.

First, let’s discuss some fundamentals of regression.

# declare paths and libraries
user = 2
if(user == 1){
    source("/home/user1/Dropbox/Medicaid_South/code/filepath.r")
}else{
    source("/Users/vshrestha/Dropbox/Medicaid_South/code/filepath.r")
}


library(pacman)
p_load(fixest, dplyr, ggplot2, tidyverse, patchwork, arrow)
theme_set(theme_minimal())