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/linregress.h>
#include <rw/analytics/ffunc.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 ( )

inline

Constructs an empty evaluation function object. Behavior is undefined.

Member Function Documentation

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.