IExogenousForecastModel<T>
Interfaces · AiDotNet.Interfaces
Marks a time-series model that forecasts from future exogenous regressors (e.g. ARIMAX, dynamic regression with ARIMA errors). The forecast horizon is the row count of the supplied matrix, and each row carries the exogenous-variable values for that future step.
Known implementations: ARIMAXModel<T>, DynamicRegressionWithARIMAErrors<T>
How It Works
For Beginners: some time-series models forecast using known future conditions —
“next month is a holiday”, “the promotion runs for N days”. Those are exogenous
variables. To forecast N steps ahead you pass an N-row matrix whose columns are those
future values. The unified Predict
front routes such models here, so callers use one Predict entry point for every
time-series model class (univariate, exogenous, multivariate).
Methods
| Method | Summary |
|---|---|
Vector<T> ForecastWithExogenous(Matrix<T> futureExogenous) |
Forecasts future target values from the future exogenous regressors. |