Getting a causal answer out of data nobody randomised. Endogeneity and its three sources, omitted variable bias with its sign predictable in advance, instrumental variables and the exclusion restriction that cannot be tested, two-stage least squares, and what a weak instrument does to it. Then the inference corrections that leave the coefficients alone: heteroskedasticity, heteroskedasticity-consistent and clustered standard errors, serial correlation, and the variance inflation factor. It closes with the designs (panel fixed effects, random effects and Hausman, difference-in-differences, regression discontinuity), the time-series traps (spurious regression, cointegration and error correction), and why a reported p value describes a test that was never run in isolation. Scoped against statistics, which owns regression, maximum likelihood, GLMs, model selection and time series basics, and introduces causal inference conceptually.
Free to start · adaptive placement finds your level · reviews timed so it stays learned.
Every idea is taught with motivation and a worked example before the drills, and an FSRS spaced-repetition engine schedules each review for the moment just before you'd forget it. A short placement check finds what you already know, so you start Econometrics exactly where it's useful.