BayesicFitting

Model Fitting and Evidence Calculation

View project on GitHub
Navigation

Global Documentation
Home Manual
Readme Glossary
Notes Design
Examples Classes
Trouble
Style



class JeffreysPrior( Prior )Source

Jeffreys prior distribution, for scale-like parameters.

Jeffreys prior is a improper prior ( i.e. its integral is unbound ).

Because of that it always needs limits, low and high, such that 0 < low < high < +Inf.

  Pr( x ) = 1.0 / ( x * norm ) if ( low < x < high ) else 0

where norm = log( high ) - log( low )

No limits are set by default.

domain2unit:
     u = ( log( d ) - log( lo ) ) / ( log( hi ) - log( lo ) ); unit2domain:
     d = exp( u * ( log( hi ) - log( lo ) ) + log( lo ) );

Examples

pr = JeffreysPrior()                       # unbound prior
pr = JeffreysPrior( limits=[0.1,1.0] )     # limited to the range [0.1,1.0]

Hidden Attributes

  • _logLo : float
         log( lowLimit )
  • _norm : float
         log( highLimit / lowLimit )

Attributes from Prior

lowLimit, highLimit, deltaP, _lowDomain, _highDomain

The default of lowLimit and _lowDomain is zero.

JeffreysPrior( limits=None, prior=None )

Default constructor.

Parameters

  • limits : list of 2 floats
         2 limits resp. low and high
  • prior : JeffreysPrior
         prior to copy (with new limits if applicable)

copy( )

getIntegral( )
Return the integral of JeffreysPrior from lowLimit to highLimit.

domain2Unit( dval )
Return a value in [0,1] given a value within the valid domain of a parameter for a Jeffreys distribution.

Parameters

  • dval : float
         value within the domain of a parameter

unit2Domain( uval )
Return a value within the valid domain of the parameter given a value between [0,1] for a Jeffreys distribution.

Parameters

  • uval : float
         value within [0,1]

result( x )
Return a the result of the distribution function at x.

Parameters

  • x : float
         value within the domain of a parameter

partialLog( p )
Return partial derivative of log( Prior ) wrt parameter.

Parameters

  • p : float
         the value

isBound( )

Return true if the integral over the prior is bound.

shortName( )

Return a string representation of the prior.

Methods inherited from Prior