//::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::
/** @author  John Miller
 *  @version 1.6
 *  @date    Tue Dec 31 14:24:47 EST 2019
 *  @see     LICENSE (MIT style license file).
 *
 *  @title   Model Framework: Forecaster for Matrix Input
 */

package scalation.analytics
package forecaster

import scalation.linalgebra.{MatriD, VectoD}

//::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::
/** The `ForecasterMat` trait add the 'forecast' method to `PredictorMat` subclasses
 *  facilitating the adaption of predictive modeling techniques such as `Regression`
 *  for use in forecasting, viz. `Regression4TS`.
 *  Note:  for forecasting, cross-validation needs to be of the rolling-validation form.
 *  @see the 'crossValidate' method in `RollingValidation`
 */
trait ForecasterMat 
{
    //::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::
    /** Produce a forecast for 'h' steps ahead into the future from current time 't'.
     *  @param xe  the relevant (optionally expanded) input/data matrix
     *  @param t   the "current" time to be forecasted, e.g., h = 1 => current, h = 2 => future
     *  @param h   the forecasting horizon, number of steps ahead to produce forecast
     */
    def forecast (xe: MatriD, t: Int, h: Int = 1): Double

    //::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::
    /** Forecast values for all time points using 1 through 'h'-steps ahead forecasts.
     *  Return a vector of forecasts for all time points.
     *  @param xe  the relevant (optionally expanded) input/data matrix
     *  @param h   the forecasting horizon, number of steps ahead to produce forecasts
     */
    def forecastAll (xe: MatriD = null, h: Int = 1): VectoD = ???

} // ForecasterMat