RegModels::LnFit Function

void LnFit(TVec B, TVec X, TVec Y, TVec Weights = null);

Fits simple logarithm equation y(x)=b[0] + b[1]*ln(x) to data.

#	Name	Type	Description
1	B	TVec *	Returns regression coefficients for natural logarithm function.
2	X	TVec *	Vector of independent variable.
3	Y	TVec *	Vector of dependent variable.
4	Weights = null	TVec *	Weights (optional). Weights are used only if they are set.

Remarks:

The routine fits equations to data by minimizing the sum of squared residuals. The observed values obey the following equation:

Y = b[0] + b[1] \ln(x)

Examples

#include "MtxExpr.h"
#include "Units.RegModels.h"

using namespace Dew;
using namespace Dew::Math;
using namespace Dew::Stats;
using namespace Dew::Stats::Units::RegModels;

void Example()
{
    Vector X(100);
    Vector Y(100);
    Vector B(0);
    Vector YHat(0);

    X.Ramp(0.5, 0.05); // x= 0.5, 0.55, ...
    Y.RandGauss(3.5, 0.12); // sample data
    // calculate coefficients
    LnFit(B,X,Y,nullptr);
    // evaluate y by using calculated coefficients
    LnEval(B,X, YHat);
}

See Also: RegModels::LnEval

Declared in Dew::Stats::Units::RegModels · Dew.Stats/Units.RegModels.h · Cross-compiler

void LnFit(TVec *B, TVec *X, TVec *Y, TVec *Weights = null);

void LnFit(TVec B, TVec X, TVec Y, TVec Weights = null);