Metrics

internal validity

whether a regression accurately estimates the causal effect we're trying to measure

Omitted Variable Bias

Leaving out a factor that affects Y and is correlated with X

Wrong Functional Form

Using the wrong mathematical relationship to explain relationship between variables

Measurement Error

Measuring/misreporting a variable incorrectly

Sample Selection Bias

The sample is selected in a way that distorts the relationship

Simultaneous Causality

X affects Y, but Y also affects X

Perfect multicollinearity

Two or more independent variables have an exact linear relationship. OLS cannot uniquely estimate all their coefficients.

Imperfect multicollinearity

Independent variables are related, but not in an exact linear relationship. OLS can still estimate their coefficients, but strong multicollinearity makes those estimates less precise.