Packages

object TranRegression extends ModelFactory

The TranRegression companion object provides transformation and inverse transformation function based on the parameter 'lambda'. It support the family of Box-Cox transformations. Note, 'rescale' is defined in ModelFactory in Model.scala.

Linear Supertypes
ModelFactory, Error, AnyRef, Any
Ordering
  1. Alphabetic
  2. By Inheritance
Inherited
  1. TranRegression
  2. ModelFactory
  3. Error
  4. AnyRef
  5. Any
  1. Hide All
  2. Show All
Visibility
  1. Public
  2. All

Value Members

  1. def allForms(x: MatriD): MatriD

    Create all forms/terms for each row/point placing them in a new matrix.

    Create all forms/terms for each row/point placing them in a new matrix.

    x

    the original un-expanded input/data matrix

    Definition Classes
    ModelFactory
  2. def apply(x: MatriD, y: VectoD, fname: Strings, hparam: HyperParameter, tran: FunctionS2S, itran: FunctionS2S, technique: RegTechnique.RegTechnique, bounds: PairD): TranRegression

    Create a TranRegression with automatic rescaling from a data matrix and response vector.

    Create a TranRegression with automatic rescaling from a data matrix and response vector.

    x

    the data/input matrix

    y

    the response/output vector

    fname

    the feature/variable names

    hparam

    the hyper-parameters (currently none)

    tran

    the transformation function (defaults to log)

    itran

    the inverse transformation function to rescale predictions to original y scale (defaults to exp)

    technique

    the technique used to solve for b in x.t*x*b = x.t*y

    bounds

    the bounds for rescaling

  3. def apply(xy: MatriD, fname: Strings, hparam: HyperParameter, tran: FunctionS2S, itran: FunctionS2S, technique: RegTechnique.RegTechnique, bounds: PairD): TranRegression

    Create a TranRegression with automatic rescaling from a combined data matrix.

    Create a TranRegression with automatic rescaling from a combined data matrix.

    xy

    the combined data/input and response/output matrix

    fname

    the feature/variable names

    hparam

    the hyper-parameters (currently none)

    tran

    the transformation function (defaults to log)

    itran

    the inverse transformation function to rescale predictions to original y scale (defaults to exp)

    technique

    the technique used to solve for b in x.t*x*b = x.t*y

    bounds

    the bounds for rescaling

  4. def apply(x: MatriD, y: VectoD, fname: Strings = null, hparam: HyperParameter = null, technique: RegTechnique.RegTechnique = QR): TranRegression

    Create a TranRegression object that uses a Box-Cox transformation.

    Create a TranRegression object that uses a Box-Cox transformation. To change 'lambda' from its default value, call 'set_lambda' first.

    x

    the data/input matrix

    y

    the response/output vector

    fname

    the feature/variable names

    hparam

    the hyper-parameters (currently none)

    technique

    the technique used to solve for b in x.t*x*b = x.t*y

  5. def box_cox(y: Double): Double

    Transform 'y' using the Box-Cox transformation.

    Transform 'y' using the Box-Cox transformation.

    y

    the value to be transformed

  6. def cox_box(z: Double): Double

    Inverse transform 'z' using the Box-Cox transformation.

    Inverse transform 'z' using the Box-Cox transformation.

    z

    the value to be inverse transformed

  7. val drp: (Null, Null, (Double) ⇒ Double, (Double) ⇒ Double, RegTechnique.Value, Null)
  8. final def flaw(method: String, message: String): Unit
    Definition Classes
    Error
  9. def forms(xi: VectoD, k: Int, nt: Int): VectoD

    Given a vector/point 'v', compute the values for all of its forms/terms, returning them as a vector (assumes Regression with intercept).

    Given a vector/point 'v', compute the values for all of its forms/terms, returning them as a vector (assumes Regression with intercept). Override for expanded columns, e.g., QuadRegression.

    xi

    the vector/point (i-th row of x) for creating forms/terms

    k

    the number of features/predictor variables (not counting intercept)

    nt

    the number of terms

    Definition Classes
    ModelFactory
  10. def numTerms(k: Int): Int

    The number of terms/parameters in the model (assumes Regression with intercept.

    The number of terms/parameters in the model (assumes Regression with intercept. Override for expanded columns, e.g., QuadRegression.

    k

    the number of features/predictor variables (not counting intercept)

    Definition Classes
    ModelFactory
  11. def rescaleOff(): Unit

    Turn rescaling off.

    Turn rescaling off.

    Definition Classes
    ModelFactory
  12. def rescaleOn(): Unit

    Turn rescaling on.

    Turn rescaling on.

    Definition Classes
    ModelFactory
  13. def setLambda(lambda_: Double): Unit

    Set the value for the 'lambda' parameter.

    Set the value for the 'lambda' parameter. Must be called before Box-Cox 'apply' method.

    lambda_

    the new value for the 'lambda' parameter