Unit root
Topic of Forecasting, part of Time series
Unit root
| Type | Topic |
| Taught in | Forecasting |
| Part | Time series |
| Questions using it | 1 |
An autoregression has a unit root when its characteristic polynomial has a root at 1, as the random walk X_t=X_{t-1}+\varepsilon_t does. The series is then not stationary, its variance grows with time, and differencing it once removes the root. Concept check C47 asks about it.
Used in
- Taught in Forecasting
- Questions that use it (1) — C47
- All 2 questions on Forecasting
See also
Autoregression and moving averages · Stationarity of a time series · Variance · ARIMA model · Accidental-deaths data · Traffic data · Partial autocorrelation · Seasonality