What to explore
Change parameters and watch the model adjust.
- True causal slope, omitted-variable effect, confounding correlation, noise, and sample size
- The notebook keeps the simulated design deterministic so the bias story is easy to isolate
Upper-undergraduate econometrics
Omitted Variable Bias
A simulated regression notebook where an unobserved confounder shifts the naive OLS slope away from the true causal effect.
Focus
Bias direction, confounding strength, and benchmark truth
Change confounding strength and the omitted variable's effect to compare the true coefficient, the naive regression, and the fully controlled benchmark.Core ideas
Interpret the mechanics before you chase the graphs.
- Bias appears when the omitted factor both affects the outcome and moves with the observed regressor.
- The sign of the bias depends on the sign pattern of the omitted effect and the confounding correlation.
- Comparing the naive and controlled regressions makes the identification problem visible rather than abstract.
Learning goals
What this model should help students internalize.
- See when a naive regression coefficient differs from the true structural effect.
- Relate the sign and magnitude of bias to correlation between the regressor and the omitted factor.
- Use the controlled regression as a benchmark for what the naive estimate is missing.
Prerequisites
Concepts to review before diving in.
- Simple OLS intuition
- Basic understanding of confounding and conditional relationships
Next models to study
Keep moving through the track.
Upper-undergraduate econometrics
Difference-in-Differences
Adjust treatment effects, group gaps, noise, and trend violations to see exactly when the DiD estimate matches the truth and when it drifts.
Upper-undergraduate econometrics
Instrumental Variables
See when IV improves on OLS, when weak instruments make the estimate unstable, and how exclusion violations undermine the design.