RWLinRegressFStatistic Class Reference

Function object that takes the data and parameters associated with a linear regression model and returns the F statistic for that model. More...

#include <rw/analytics/ffunc.h>

Inherits RWModelEvalFunctor< T, S >.

Public Member Functions
	RWLinRegressFStatistic ()
virtual double	operator() (const RWGenMat< double > &xdata, const RWMathVec< double > &ydata, const RWMathVec< double > &coefficients) const

Detailed Description

Class RWLinRegressFStatistic is a function object that takes the data and parameters associated with a linear regression model and returns the F statistic for that model. An equation for the F statistic can be found in Section 3.2.4, "Significance of the Model (Overall F Statistic)," in the Business Analysis Module User's Guide.

This class is useful in conjunction with the model selection class RWLinRegModelSelector, which requires a function object to evaluate the quality of different predictor variable subsets. However, it can also be used by itself as a direct way of calculating the F statistic.

Synopsis

#include <rw/analytics/ffunc.h>
 
RWLinRegressFStatistic fstat;

Example

#include <rw/analytics/ffunc.h>
#include <rw/analytics/linregress.h>
 
int main() {
  RWGenMat<double> predictorMatrix =
      "5x2 [1.2 2.1 8 7 3 3.2 6.4 4.6 2 2.3]";
  RWMathVec<double> observationVector = "[2.5 3.7 1.4 2.3 5.6]";
 
  RWLinearRegression lr(predictorMatrix, observationVector);
  RWLinRegressFStatistic fstat;
 
  // Print the F statistic value if parameter
  // calculation succeeded.
  if (!lr.fail()) {
    cout << "F statistic: "
         << fstat(lr.regressionMatrix(), lr.observationVector(),
                  lr.parameters())
         << endl;
  } else {
    cout << "Parameter calculation failed." << endl;
  }
  return 0;
}

Constructor & Destructor Documentation

RWLinRegressFStatistic()

RWLinRegressFStatistic::RWLinRegressFStatistic ( )

inline

Constructs an empty evaluation function object. Behavior is undefined.

Member Function Documentation

operator()()

virtual double RWLinRegressFStatistic::operator() ( const RWGenMat< double > & xdata, const RWMathVec< double > & ydata, const RWMathVec< double > & coefficients ) const

virtual

Given a regression class instance, this function returns the F statistic associated with the predictor variables selected in the model.