RWLogisticRegressionParam Class Reference
[Parameter Calculation]

Container class for logistic regression parameter estimates and their associated statistical quantities. More...

#include <rw/analytics/logregparam.h>

List of all members.

Public Member Functions
	RWLogisticRegressionParam ()
	RWLogisticRegressionParam (const RWLogisticRegressionParam &rhs)
double	standardError () const
double	value () const
double	waldChiSqStatistic () const
double	waldChiSqStatPValue () const
double	waldChiSqStatCriticalValue (double alpha) const
RWInterval< double >	confidenceInterval (double alpha) const
RWLogisticRegressionParam &	operator= (const RWLogisticRegressionParam &rhs)

---

Detailed Description

RWLogisticRegressionParam is the container class for logistic regression parameter estimates and their associated statistical quantities. The estimates are described in the Business Analysis Module User's Guide.

Synopsis

 #include <rw/analytics/logregress.h>
 #include <rw/analytics/logparam.h>

 RWLogisticRegression lr;
 RWTValVector<RWLogisticRegressionParam> p = lr.parameterEstimates();

Examples

 #include <rw/analytics/logregress.h>
 #include <iostream>

 int main()
 {
  RWGenMat<double> predData = "5x2
                              [1 234 2 431 3 333 4 654 5 788]";
  RWMathVec<bool> obsData(5, rwUninitialized );
  obsData[0] = obsData[3] = obsData[4] = true;
  obsData[1] = obsData[2] = false;

 RWLogisticRegression lr( predData, obsData );

   // Make sure parameter calculation succeeded.
   if ( lr.fail() )
   {
     return 0;
   }

   double sigLevel = .05;

   // Print out model parameter estimate info.
   RWTValVector<RWLogisticRegressionParam> params =
                                lr.parameterEstimates();
   for ( size_t i = 0; i < params.length(); i++ )
   {
     std::cout << "Model parameter " << i
               << (i==0 ? " Intercept:" : ":") << std::endl;
     std::cout << "  value:                      "
               << params[i].value() << std::endl;
     std::cout << "  standard error:             "
               << params[i].standardError() << std::endl;
     std::cout << "  Wald statistic:                "
               << params[i].waldChiSqStatistic() << std::endl;
     std::cout << "  Wald statistic P-value:        "
               << params[i].waldChiSqStatPValue() << std::endl;
     std::cout << "  Wald statistic critical value: "
               <<  params[i].waldChiSqStatCriticalValue(sigLevel)
               <<  std::endl;
     std::cout << "  " << sigLevel << " confidence interval:  " << "["
               << params[i].confidenceInterval(sigLevel).lowerBound()
               << ", "
               << params[i].confidenceInterval(sigLevel).upperBound()
               << "]" << std::endl;
   }

   return 0;
 }

---

Constructor & Destructor Documentation

RWLogisticRegressionParam::RWLogisticRegressionParam

(

)

[inline]

Constructs an empty fitted parameter object. Behavior undefined.

RWLogisticRegressionParam::RWLogisticRegressionParam

(

const RWLogisticRegressionParam &

rhs

)

[inline]

Constructs a copy of rhs.

---

Member Function Documentation

RWInterval<double> RWLogisticRegressionParam::confidenceInterval

(

double

alpha

)

const

Returns an alpha level confidence interval for the parameter.

RWLogisticRegressionParam& RWLogisticRegressionParam::operator=

(

const RWLogisticRegressionParam &

rhs

)

Assignment operator.

double RWLogisticRegressionParam::standardError

(

)

const

Returns the estimated standard error for the fitted value. This is the square root of the estimated variance, V, described in the Business Analysis Module User's Guide.

double RWLogisticRegressionParam::value

(

)

const [inline]

Returns the least squares estimate for the parameter.

double RWLogisticRegressionParam::waldChiSqStatCriticalValue

(

double

alpha

)

const [inline]

Returns the critical value for the Wald chi-square statistic at significance level alpha.

double RWLogisticRegressionParam::waldChiSqStatistic

(

)

const [inline]

Returns the Wald chi-square statistic for the hypothesis that the parameter is equal to 0.

double RWLogisticRegressionParam::waldChiSqStatPValue

(

)

const [inline]

Returns the P-value for the parameter Wald chi-square statistic.