gov.sandia.cognition.statistics
Interface ProbabilityFunction<DataType>

Type Parameters:
DataType - Type of data that can be sampled from the distribution, and also the input to the evaluate method.
All Superinterfaces:
Cloneable, CloneableSerializable, ComputableDistribution<DataType>, Distribution<DataType>, Evaluator<DataType,Double>, Serializable
All Known Subinterfaces:
DataDistribution.PMF<KeyType>, ProbabilityDensityFunction<DataType>, ProbabilityMassFunction<DataType>, UnivariateProbabilityDensityFunction
All Known Implementing Classes:
AdaptiveRejectionSampling.UpperEnvelope, BernoulliDistribution.PMF, BetaBinomialDistribution.PMF, BetaDistribution.PDF, BinomialDistribution.PMF, CategoricalDistribution.PMF, CauchyDistribution.PDF, ChineseRestaurantProcess.PMF, ChiSquareDistribution.PDF, DefaultDataDistribution.PMF, DeterministicDistribution.PMF, DirichletDistribution.PDF, ExponentialDistribution.PDF, GammaDistribution.PDF, GeometricDistribution.PMF, InverseGammaDistribution.PDF, InverseWishartDistribution.PDF, LaplaceDistribution.PDF, LogisticDistribution.PDF, LogNormalDistribution.PDF, MixtureOfGaussians.PDF, MultinomialDistribution.PMF, MultivariateGaussian.PDF, MultivariateMixtureDensityModel.PDF, MultivariatePolyaDistribution.PMF, MultivariateStudentTDistribution.PDF, NegativeBinomialDistribution.PMF, NormalInverseGammaDistribution.PDF, NormalInverseWishartDistribution.PDF, ParetoDistribution.PDF, PoissonDistribution.PMF, ScalarDataDistribution.PMF, ScalarMixtureDensityModel.PDF, StudentTDistribution.PDF, UniformDistribution.PDF, UnivariateGaussian.PDF, WeibullDistribution.PDF, YuleSimonDistribution.PMF

public interface ProbabilityFunction<DataType>
extends ComputableDistribution<DataType>, Evaluator<DataType,Double>

A Distribution that has an evaluate method that indicates p(x), such as a probability density function or a probability mass function (but NOT a cumulative distribution function).

Since:
3.0
Author:
Kevin R. Dixon

Method Summary
 double logEvaluate(DataType input)
          Evaluate the natural logarithm of the distribution function.
 
Methods inherited from interface gov.sandia.cognition.statistics.ComputableDistribution
getProbabilityFunction
 
Methods inherited from interface gov.sandia.cognition.statistics.Distribution
sample, sample
 
Methods inherited from interface gov.sandia.cognition.util.CloneableSerializable
clone
 
Methods inherited from interface gov.sandia.cognition.evaluator.Evaluator
evaluate
 

Method Detail

logEvaluate

double logEvaluate(DataType input)
Evaluate the natural logarithm of the distribution function. This is sometimes more efficient than evaluating the distribution function itself, and when evaluating the product of many independent or exchangeable samples.

Parameters:
input -
Returns:
Natural logarithm of the distribution function.