1 · Targets, transformations, and stationarity
Define the forecasting target before choosing a model.
Univariate Time Series & Forecasting
Stationarity, ACF/PACF, ARMA and ARIMA modeling, diagnostics, and honest rolling-origin forecast evaluation.
Course overview
Learn Univariate Time Series & Forecasting interactively
This free course uses real Python and statsmodels in your browser. Work through the modules below with explanations, executable examples, interactive controls, and questions.
2 · ACF, PACF, and dependence
Read sample correlation patterns as evidence, not an automatic order selector.
3 · ARMA dynamics
Combine persistent states with transitory innovations.
4 · ARIMA and seasonal structure
Difference parsimoniously and model the remaining dynamics.
5 · Rolling-origin forecast evaluation
Judge forecasts using information that was genuinely available at each origin.