Pulsar::PolnProfileFit::Analysis Class Reference

Analysis of the matrix template matching algorithm. More...

#include <PolnProfileFitAnalysis.h>

Public Member Functions
	Analysis ()
	Default constructor.
void	set_compute_error (bool flag=true)
	When set, estimate the uncertainty in each attribute.
void	set_fit (PolnProfileFit *)
	Set the PolnProfileFit algorithm to be analysed.
Estimate< double >	get_relative_error () const
	Get the relative arrival time error.
Estimate< double >	get_multiple_correlation () const
	Get the multiple correlation between phase shift and Jones parameters.
Estimate< double >	get_relative_conditional_error () const
	Get the relative conditional arrival time error.
double	get_c_varphi ()
	Get the variance of varphi.
double	get_C_varphi ()
	Get the variance of varphi.
double	get_expected_relative_error (std::vector< unsigned > &histogram)
	Get the expected relative conditional error given a histogram.
double	get_c_varphi_error () const
	Get the uncertainty in the variance of varphi.
double	get_Rmult ()
	Get the multiple correlation.
double	get_cond_var ()
	Get the conditional variance of varphi.
Matrix< 8, 8, double >	get_correlation () const
	Get the correlation coefficients.
void	output_C_varphi (const char *filename)
	Write the terms in C_varhpi out to file.

Static Public Attributes
static bool	verbose = false
	Verbosity flag.

Protected Member Functions
void	build (bool only_relative_error=false)
	Computes the above three numbers.
Jones< double >	del_deleta (unsigned i, const Jones< double > &K) const
	The partial derivative of K with respect to free parameter, eta.
double	delR2_varphiJ_delS (Matrix< 8, 8, double > &delC_delS)
	The partial derivative of the multiple correlation squared wrt S_k.
void	delC_delS (Matrix< 8, 8, double > &delC_delSre, Matrix< 8, 8, double > &delC_delSim, unsigned k) const
	The partial derivative of the covariance matrix wrt Re[S_k] and Im[S_k].
Jones< double >	delrho_delS (unsigned k) const
	The partial derivative of rho wrt Stokes parameter.
void	get_curvature (Matrix< 8, 8, double > &curvature)
	Get the curvature matrix.
void	add_curvature (Matrix< 8, 8, double > &curvature)
void	set_harmonic (unsigned index)
	Set the model up to evaluate the specified harmonic.
void	evaluate (unsigned index)
	Evaluate the model and phase (and gradients) of the specified harmonic.

Protected Attributes
Estimate< double >	relative_error
	The relative arrival time error.
Estimate< double >	relative_conditional_error
	The relative conditional arrival time error.
Estimate< double >	multiple_correlation
	The multiple correlation between phase shift and Jones parameters.
bool	built
	Flag set true when all three of the above values have been computed.
Reference::To< PolnProfileFit >	fit
	The PolnProfileFit algorithm to be analysed.
Reference::To< MEAL::Complex2 >	xform
	The transformation built into the MTM algorithm.
unsigned	max_harmonic
	The maximum harmonic used when computing the optimal basis.
Jones< double >	model_result
std::vector< Jones< double > >	model_gradient
Jones< double >	xform_result
std::vector< Jones< double > >	xform_gradient
Jones< double >	phase_result
std::vector< Jones< double > >	phase_gradient
MEAL::StokesError	error
	Propagation of uncertainty through the congruence tranformation.
MEAL::StokesCovariance	stokes_covariance
	Propagation of Stokes parameter covariances through congruence xform.
Matrix< 8, 8, double >	covariance
	the covariance matrix
Matrix< 7, 7, double >	C_JJ
	the covariances between the Jones matrix parameters
Matrix< 7, 7, double >	inv_C_JJ
	the inverse of the covariances between the Jones matrix parameters
Vector< 7, double >	C_varphiJ
	the covariances between the phase shift and each Jones matrix parameter
double	c_varphi
	the variance of the unconditional phase shift
double	R2_varphiJ
	the multiple correlation squared
double	efac
bool	compute_error
	Compute the uncertainty of results.
std::vector< Reference::To< MEAL::Complex2 > >	inputs
	The original inputs.
Matrix< 4, 4, double >	Xi
	Covariances between the four Stokes parameters.

Detailed Description

Analysis of the matrix template matching algorithm.

Member Function Documentation

delR2_varphiJ_delS()

double Pulsar::PolnProfileFit::Analysis::delR2_varphiJ_delS ( Matrix< 8, 8, double > & delC_delS )

protected

The partial derivative of the multiple correlation squared wrt S_k.

Return the partial derivative of the multiple correlation squared with respect to the real or imaginary component of a Stokes parameter

References c_varphi, C_varphiJ, delC_delS(), inv_C_JJ, and R2_varphiJ.

Referenced by build().

get_curvature()

void Pulsar::PolnProfileFit::Analysis::get_curvature ( Matrix< 8, 8, double > & alpha )

protected

Get the curvature matrix.

Computes Equation 14 of van Straten (2006)

References evaluate(), fit, max_harmonic, and verbose.

Referenced by build(), get_c_varphi(), and get_Rmult().

---

The documentation for this class was generated from the following files:

• PolnProfileFitAnalysis.h
• PolnProfileFitAnalysis.C