• Home
• com.opengamma.strata
• strata-math
• 2.12.38
• source code
• NonLinearParameterTransforms.java

×

Project download

Many resources are needed to download a project. Please understand that we have to compensate our server costs. Thank you in advance.
Project price only 1 $

You can buy this project and download/modify it how often you want.

com.opengamma.strata.math.impl.minimization.NonLinearParameterTransforms Maven / Gradle / Ivy

/*
 * Copyright (C) 2011 - present by OpenGamma Inc. and the OpenGamma group of companies
 *
 * Please see distribution for license.
 */
package com.opengamma.strata.math.impl.minimization;

import com.opengamma.strata.collect.array.DoubleArray;
import com.opengamma.strata.collect.array.DoubleMatrix;

/**
 * Describes the transformation (and its inverse) from  a set of n variables (e.g. model parameters) to a set of m variables
 * (e.g. fitting parameters), where m <= n.The principle use is in constrained optimisation, where the valid values of the parameters
 * of a model live in some hyper-volume in R^n, but we wish to work with unconstrained variables in R^m. 

 * The model parameters are denoted as y and the unconstrained variables as y*, which are related by the vector function
 * y* = f(y), and its inverse  y = f-1(y*). The i,j element of the Jacobian is the rate of change of 
 * the ith element of y* with respect to the  j th element of y, which is a (matrix) function
 * of y, i.e. J(y). The inverse Jacobian is the rate of change of y with respect to 
 * y*, i.e. J-1(y*).
 * These four functions must be provided by implementations of this interface. 
 */
//CSOFF: JavadocMethod
public interface NonLinearParameterTransforms {

  public abstract int getNumberOfModelParameters();

  public abstract int getNumberOfFittingParameters();

  /**
   * Transforms from a set of model parameters to a (possibly smaller) set of unconstrained fitting parameters.
   * @param modelParameters   the model parameters
   * @return The fitting parameters
   */
  public abstract DoubleArray transform(DoubleArray modelParameters);

  /**
   * Transforms from a set of unconstrained fitting parameters to a (possibly larger) set of function parameters.
   * @param fittingParameters The fitting parameters
   * @return The model parameters
   */
  public abstract DoubleArray inverseTransform(DoubleArray fittingParameters);

  /**
   * Calculates the Jacobian - the rate of change of the fitting parameters WRT the model parameters.
   * @param modelParameters The model parameters 
   * @return The Jacobian 
   */
  public abstract DoubleMatrix jacobian(DoubleArray modelParameters);

  /**
   * Calculates the inverse Jacobian  - the rate of change of the model parameters WRT the fitting parameters.
   * @param fittingParameters The fitting parameters
   * @return the inverse Jacobian 
   */
  public abstract DoubleMatrix inverseJacobian(DoubleArray fittingParameters);

}