Skip to content
~/patrick/projects/dags-causal-inference
← projects

model lane

DAGs and Causal Inference

A project integrating directed acyclic graphs (DAGs) with conventional econometric techniques, covering collider bias, the front-door criterion, M-bias, and falsification testing.

pipeline
  1. data (not covered)
  2. model (covered)
  3. eval (covered)
  4. serve (not covered)
  5. monitor (not covered)
stack

EconometricsCausal inferenceDAGsOLSInstrumental variablesMonte Carlo

links

write-up / live ↗

This project uses Monte Carlo simulations to show where a standard regression goes wrong, like when you adjust for a collider or run into M-bias, and how reasoning with a DAG helps you choose which variables to adjust for.

It also covers falsification testing, where you check a DAG against the data by evaluating the conditional independencies that the DAG implies.

The Spot the Confounder game in the lab is built on the same ideas.

nextESOP Tax Credit, a Difference-in-Differences Analysis →

keyboard

⌘K /
Search and commands
t T
Next / previous theme
j k
Move through items on the page (k on the first one goes back to the top)
o
Open the selected item
g then h w p b a l
Home, Work, Projects, Blog, About, Lab
c
Copy email
?
This list

Choose a theme

Dark

Light

You can also press t to step through them.