11.3 Roadmap

Rearview Mirror

• Statisticians create a population model to represent the world.
• The BLP is a useful way to summarize relationships in a model, and OLS regression is a way to estimate the BLP.
• OLS regression is a foundational tool that can be applied to questions of description

Today

• Questions of explanation require a substantially different modeling process.
• To answer causal questions, we must work within a causal theory
• OLS regression is sometimes appropriate for measuring a causal effect,
• But, only when the model estimated matches the causal theory.
• So, we must watch out for omitted variable bias, reverse causality, and outcome variables on the right hand side.

Looking Ahead

• The famous Classical Linear Model (CLM) allows us to apply regression to smaller samples.
• We will address the pervasive issue of false discovery, and ways to be a responsible member of the scientific community.