Packages

object ForecasterVec

The ForecasterVec companion object provides functions useful for forecasting models. It also contains a sample dataset.

Linear Supertypes
AnyRef, Any
Ordering
  1. Alphabetic
  2. By Inheritance
Inherited
  1. ForecasterVec
  2. AnyRef
  3. Any
  1. Hide All
  2. Show All
Visibility
  1. Public
  2. All

Value Members

  1. val MAX_LAGS: Int
  2. def differ(u: VectoD, v: VectoD, scale: Double = 1E-9, allow: Boolean = true): Int

    Point out the differences between two vectors/time-series.

    Point out the differences between two vectors/time-series.

    u

    the first vector/time-series

    v

    the second vector/time-series

    scale

    the scale factor to set the tolerance 'tol'

    allow

    flag indicating whether allow (via assert) any differences

  3. def durbinLevinson(g: VectoD, ml: Int): MatriD

    Apply the Durbin-Levinson Algorithm to iteratively compute the 'psi' matrix.

    Apply the Durbin-Levinson Algorithm to iteratively compute the 'psi' matrix. The last row of the matrix gives 'AR' coefficients.

    g

    the auto-covariance vector (gamma)

    ml

    the maximum number of lags

    See also

    www.stat.tamu.edu/~suhasini/teaching673/time_series.pdf, p. 247

  4. val t: VectorD
  5. def test(modName: String, p: Int, mod: ForecasterVec, kt: Int = 5, h: Int = 1): Unit

    Test forecasting model that extends ForecasterVec.

    Test forecasting model that extends ForecasterVec.

    modName

    the name of the model to test

    p

    the order of the model (number of response values to use in forecasts)

    mod

    the model to test

    kt

    the retraining frequency

    h

    the forecasting horizon

  6. def testInSamp(y: VectoD, pq: Int, mod: ForecasterVec, h: Int = 1): MatriD

    Test (in sample) forecasting model that extends ForecasterVec.

    Test (in sample) forecasting model that extends ForecasterVec. returning the matrix of all forecasts over horizon and time.

    y

    the time series data

    pq

    the order of the model (number of response values to use in forecasts)

    mod

    the model to test

    h

    the forecasting horizon

  7. val y: VectorD