Upper-undergraduate econometrics

Instrumental Variables

A simulated endogeneity notebook comparing OLS and IV under changes in instrument strength, endogeneity, and exclusion validity.

Econometrics Causal inference Advanced EasyEcon / Marimo Regression, identification, and simulated evidence
Focus

Endogeneity, relevance, exclusion, and weak instruments

See when IV improves on OLS, when weak instruments make the estimate unstable, and how exclusion violations undermine the design.

EasyEcon interactive

IV notebook

EasyEcon / Marimo

On phones, the full-screen view gives the cleanest notebook reading experience.

Open full screen

Deferred notebook load

Endogeneity, relevance, exclusion, and weak instruments

See when IV improves on OLS, when weak instruments make the estimate unstable, and how exclusion violations undermine the design.

Open full screen
The EasyEcon Marimo/WASM bundle loads only when requested or when the panel scrolls near view on fast connections.

Econometrics and causal inference

Step 5 of 5

What to explore

Change parameters and watch the model adjust.

  • True treatment effect, instrument strength, endogeneity, exclusion violation, and sample size
  • The notebook shows the first stage and the outcome relation side by side

Core ideas

Interpret the mechanics before you chase the graphs.

  • An instrument must move the endogenous regressor and stay out of the outcome except through that channel.
  • IV can correct OLS bias when the instrument is valid, but weak first stages reduce precision.
  • Exclusion violations can make IV look authoritative while still being wrong.

Learning goals

What this model should help students internalize.

  • Compare OLS and IV when the regressor is endogenous.
  • Interpret the relevance and exclusion conditions as distinct design requirements.
  • See why weak instruments make IV noisy even when the design is conceptually correct.

Prerequisites

Concepts to review before diving in.

  • Omitted-variable-bias intuition
  • Comfort with the idea of an external source of variation

Next models to study

Keep moving through the track.

Econometrics Regression Intermediate EasyEcon / Marimo

Upper-undergraduate econometrics

Simple OLS Regression

Adjust the true intercept, slope, noise, and sample size to relate realised estimates to fit, slope estimation error, and an approximate 95% slope interval.

Fitted line, residuals, precision, and known truth Open model
Econometrics Causal inference Advanced EasyEcon / Marimo

Upper-undergraduate econometrics

Omitted Variable Bias

Change confounding strength and the omitted variable's effect to compare the true coefficient, the naive regression, and the fully controlled benchmark.

Bias direction, confounding strength, and benchmark truth Open model