GaussianApprox.h File Reference

Utilities for summarizing observable samples with Gaussian approximations. More...

#include <vector>
#include <stdexcept>
#include "Statistics.h"

Include dependency graph for GaussianApprox.h:

This graph shows which files directly or indirectly include this file:

Go to the source code of this file.

Classes
struct	GaussianSummary
	Summary statistics for one binned observable distribution. More...

Functions
std::ostream &	operator<< (std::ostream &os, const GaussianSummary &gs)
	Streams a human-readable Gaussian summary.
std::vector< GaussianSummary >	gaussian_fit (const ObsSamples &S, double skew_abs_threshold=0.2)
	Builds Gaussian or split-Gaussian summaries for observable samples.

Detailed Description

Utilities for summarizing observable samples with Gaussian approximations.

This header provides a lightweight summary structure and a helper that fits a Gaussian or split-Gaussian approximation to each observable column, depending on the measured skewness.

See also

Statistics.h

Definition in file GaussianApprox.h.

Function Documentation

gaussian_fit()

std::vector< GaussianSummary > gaussian_fit ( const ObsSamples &  S, double  skew_abs_threshold = 0.2  )

inline

Builds Gaussian or split-Gaussian summaries for observable samples.

Each observable column is summarized through summarize_columns_obs. A column is marked as symmetric when the absolute skewness is below skew_abs_threshold; otherwise, the asymmetric mode and left/right standard deviations are kept for reporting.

Parameters

S	Observable samples to summarize.
skew_abs_threshold	Maximum absolute skewness accepted for the symmetric Gaussian approximation.

Returns

One GaussianSummary per observable column.

Definition at line 75 of file GaussianApprox.h.

operator<<()

std::ostream & operator<< ( std::ostream &  os, const GaussianSummary &  gs  )

inline

Streams a human-readable Gaussian summary.

Symmetric summaries are printed as mu +- sigma; asymmetric summaries are printed as mode + sigma_p - sigma_m. In both cases, the skewness is included for diagnostics.

Parameters

os	Output stream.
gs	Summary to print.

Returns

Reference to os.

Definition at line 51 of file GaussianApprox.h.