1#ifndef __JPOINT4DREGRESSOR__
2#define __JPOINT4DREGRESSOR__
51 template<
class JHit_t>
56 const double dt =
hit.getT() -
vx.getT(
hit.getPosition());
58 const double u =
dt / sigma;
60 return estimator->getRho(
u) *
hit.getW();
88 template<
class JHit_t>
93 const double dt =
hit.getT() -
vx.getT(
hit.getPosition());
95 const double u =
dt / sigma;
97 result.chi2 = estimator->getRho(
u) *
hit.getW();
101 double weight = getIndexOfRefraction() * getInverseSpeedOfLight()/(d.
getLength() * sigma);
103 result.gradient =
hit.getW() *
JPoint4D( d * weight,
105 result.gradient.mul(0.5 * estimator->getPsi(
u));
Maximum likelihood estimator (M-estimators).
General purpose messaging.
General purpose data regression method.
Fit method based on the Levenberg-Marquardt method.
Data structure for vertex fit.
Simple fit method based on Powell's algorithm, see reference: Numerical Recipes in C++,...
Data structure for vector in three dimensions.
double getLength() const
Get length.
The template JSharedPointer class can be used to share a pointer to an object.
Template definition of a multi-dimensional oscillation probability interpolation table.
Auxiliary classes and methods for linear and iterative data regression.
This name space includes all other name spaces (except KM3NETDAQ, KM3NET and ANTARES).
Abstract class for global fit method.
Data structure for return value of fit function.
result_type operator()(const JPoint4D &vx, const JHit_t &hit) const
JLANG::JSharedPointer< JMEstimator > estimator
M-Estimator function.
JRegressor(double sigma)
Constructor.
double sigma
Time resolution [ns].
double sigma
Time resolution [ns].
JLANG::JSharedPointer< JMEstimator > estimator
M-Estimator function
double operator()(const JPoint4D &vx, const JHit_t &hit) const
JRegressor(double sigma)
Constructor.
Template definition of a data regressor of given model.
Auxiliary class to set-up Hit.