Time Series Models: Smoothing, AR, and ARCH

Free SOA Exam SRM (Statistics for Risk Modeling) lesson in Time Series Models. 24 min read, ~3,658 words.

Exponential smoothing forecasts via a weighted average of past values, with weights decaying geometrically by smoothing constant. Larger reacts faster; smaller is more stable. An AR(p) model regresses on its own p most recent lags plus white noise. Stationarity requires the AR roots to lie outside the unit circle. ARCH(q)...