GraphSysErr  0.10-2
A class to hold results with statistical and systematic errors A class to hold results with statistical and systematic errors
Public Member Functions | Static Public Member Functions | List of all members
GraphSysErr::LinearSigmaCombiner Struct Reference

A combiner that uses a linear $\sigma$ approximation. More...

Inheritance diagram for GraphSysErr::LinearSigmaCombiner:
Inheritance graph
[legend]
Collaboration diagram for GraphSysErr::LinearSigmaCombiner:
Collaboration graph
[legend]

Public Member Functions

Double_t W (const Observation &r) const
 Calculate the weight. More...
 
Double_t StepW (Double_t guess, const Observation &r) const
 Calculate the weight based on a guess of best $ x'$. More...
 
Double_t StepOffset (Double_t, const Observation &) const
 Calculate the bias. More...
 
Double_t VarTerm (Double_t guess, const Observation &r) const
 Calculate the contribution variance to the $\chi^2$ with the guess $x'$. More...
 
Wrapper_t Wrapper () const
 Return function pointer to wrapper. More...
 
 ClassDef (LinearSigmaCombiner, 1)
 
- Public Member Functions inherited from GraphSysErr::Combiner
 Combiner ()
 Constructor. More...
 
virtual ~Combiner ()
 Virtual destructor. More...
 
Double_t ChiTerm (Double_t guess, const Observation &r) const
 Calculate the contribution variance to the $\chi^2$ with the guess $ x'$. More...
 
Double_t F (Double_t guess, Double_t chi2) const
 Calculate the $ \chi^2(x')$ where $ x'$ is current guess at the observation. More...
 
Double_t E (UShort_t nIter, Int_t sign, Double_t best, Double_t chi2, Double_t s)
 Try to find best error. More...
 
Double_t X (UShort_t nIter, Double_t lowest, Double_t highest)
 Find best estimate of $ x$. More...
 
Result * Calculate (UShort_t nIter=50)
 Do the calculation. More...
 
TF1 * MakeF (const Observation &r, Int_t j) const
 Make a function that represents to Log-likehood for a given observation. More...
 
TLine * MakeL (TF1 *f) const
 Make a line that represents the best found errors. More...
 
void Draw (Option_t *option="")
 
 ClassDef (Combiner, 1)
 
void Clear (Option_t *option="")
 Clear the internal data. More...
 
void Add (const Observation &r)
 Add an obervation. More...
 
void Add (Double_t x, Double_t el, Double_t eh)
 Add an observation. More...
 
void Print (Option_t *option="") const
 Print content of the list. More...
 

Static Public Member Functions

static Double_t L (Double_t guess, Double_t x, Double_t el, Double_t eh)
 Return the likely-hood function value at $ x'$: More...
 
static Double_t WrapL (Double_t *xp, Double_t *pp)
 Wrap likely-hood function for ROOT. More...
 

Additional Inherited Members

- Public Types inherited from GraphSysErr::Combiner
typedef Double_t(* Wrapper_t) (Double_t *, Double_t *)
 
- Public Attributes inherited from GraphSysErr::Combiner
TClonesArray fData
 
Result * fResult
 

Detailed Description

A combiner that uses a linear $\sigma$ approximation.

Examples:
TestCombiner.C.

Definition at line 5407 of file GraphSysErr.C.

Member Function Documentation

GraphSysErr::LinearSigmaCombiner::ClassDef ( LinearSigmaCombiner  ,
1   
)
static Double_t GraphSysErr::LinearSigmaCombiner::L ( Double_t  guess,
Double_t  x,
Double_t  el,
Double_t  eh 
)
inlinestatic

Return the likely-hood function value at $ x'$:

\[ L(x') = \left[(x'-x) / (s + s'(x'-x))\right]^2 \]

where

\[ s = 2\sigma^+\sigma^-/(\sigma^++\sigma^-) \]

\[ s' = (\sigma^+-\sigma^-)/(\sigma^++\sigma^-) \]

Parameters
guess$ x'$
x$ x$
el$ \sigma^-$
eh$ \sigma^+$
Returns

Definition at line 5495 of file GraphSysErr.C.

References GraphSysErr::Combiner::Observation::Svar().

Double_t GraphSysErr::LinearSigmaCombiner::StepOffset ( Double_t  ,
const Observation &   
) const
inlinevirtual

Calculate the bias.

Returns
0

Implements GraphSysErr::Combiner.

Definition at line 5452 of file GraphSysErr.C.

Double_t GraphSysErr::LinearSigmaCombiner::StepW ( Double_t  guess,
const Observation &  r 
) const
inlinevirtual

Calculate the weight based on a guess of best $ x'$.

\[ w(x') = s / [s + s' (x' - x)]^3 \]

Parameters
guessCurrent guess $ x'$
rObservation
Returns
$ W(x')$

Implements GraphSysErr::Combiner.

Definition at line 5440 of file GraphSysErr.C.

References GraphSysErr::Combiner::Observation::S(), and GraphSysErr::Combiner::Observation::Svar().

Double_t GraphSysErr::LinearSigmaCombiner::VarTerm ( Double_t  guess,
const Observation &  r 
) const
inlinevirtual

Calculate the contribution variance to the $\chi^2$ with the guess $x'$.

\[ v(x') = [s + s' (x' - x)]^2 \]

Parameters
guessCurrent guess $ x'$
rObservation
Returns
$ v(x')$

Implements GraphSysErr::Combiner.

Definition at line 5469 of file GraphSysErr.C.

References GraphSysErr::Combiner::Observation::Svar().

Double_t GraphSysErr::LinearSigmaCombiner::W ( const Observation &  r) const
inlinevirtual

Calculate the weight.

\[ w = 1/2 (s + x s')^3 / s \]

Parameters
rObservation
Returns
$ W$

Implements GraphSysErr::Combiner.

Definition at line 5420 of file GraphSysErr.C.

References GraphSysErr::Combiner::StepW().

static Double_t GraphSysErr::LinearSigmaCombiner::WrapL ( Double_t *  xp,
Double_t *  pp 
)
inlinestatic

Wrap likely-hood function for ROOT.

Parameters
xpPointer to independent variables
ppPointer to parameters
Returns
Likely-hood function evaluate at xp

Definition at line 5513 of file GraphSysErr.C.

Wrapper_t GraphSysErr::LinearSigmaCombiner::Wrapper ( ) const
inlinevirtual

Return function pointer to wrapper.

Returns
Function pointer

Implements GraphSysErr::Combiner.

Definition at line 5522 of file GraphSysErr.C.

References GraphSysErr::Combiner::Observation::ClassDef().

The documentation for this struct was generated from the following file:
Generated on Fri Mar 11 2016 14:59:09 for GraphSysErr by   doxygen 1.8.11