econometrics · Async + labs
Time Series Econometrics
Stationarity tests, ARIMA intuition, and forecasting intervals without overselling precision.
ZAR 1 240 · 7 weeks
Software: R · Method focus: time series · Level: intermediate · Assessment: forecast report
Description
Build intuition for unit roots, cointegration basics, and forecast evaluation with rolling windows. Emphasis on reporting uncertainty and model comparison without p-hacking.
What is included
- Rolling window forecasting notebooks
- Diebold–Mariano style comparison exercises
- Seasonality decomposition walkthroughs
- Mentor review of forecast fan charts
- Bibliography on macro data revisions
- Replication of a classic applied paper appendix
Outcomes
- Choose differencing vs deterministic trends with justification
- Communicate forecast intervals to non-technical readers
- Compare models with honest out-of-sample metrics
Dr. Chipo Ndlovu
Monetary policy forecaster turned educator; focuses on honest intervals.
Reviews
Fan chart feedback was blunt in a useful way — my team stopped showing a false sense of precision.
Rolling evaluation section changed how we report model changes quarter to quarter.
Questions
We introduce VAR notation and IRFs conceptually; deep SVAR identification is not included.
Comfort with lag operators helps; we provide refresher notes.
We use public macro series; bringing proprietary data is optional with permissions.