gov.sandia.cognition.statistics.distribution
Class StudentTDistribution.MaximumLikelihoodEstimator

java.lang.Object
  extended by gov.sandia.cognition.util.AbstractCloneableSerializable
      extended by gov.sandia.cognition.statistics.distribution.StudentTDistribution.MaximumLikelihoodEstimator
All Implemented Interfaces:
BatchLearner<Collection<? extends Double>,StudentTDistribution>, DistributionEstimator<Double,StudentTDistribution>, CloneableSerializable, Serializable, Cloneable
Enclosing class:
StudentTDistribution

public static class StudentTDistribution.MaximumLikelihoodEstimator
extends AbstractCloneableSerializable
implements DistributionEstimator<Double,StudentTDistribution>

Estimates the parameters of the Student-t distribution from the given data, where the degrees of freedom are estimated from the Kurtosis of the sample data.

See Also:
Serialized Form

Field Summary
static double DEFAULT_VARIANCE
          Typical value of a defaultVariance, 1.0E-5
 
Constructor Summary
StudentTDistribution.MaximumLikelihoodEstimator()
          Default constructor
StudentTDistribution.MaximumLikelihoodEstimator(double defaultVariance)
          Creates a new instance of MaximumLikelihoodEstimator
 
Method Summary
 StudentTDistribution learn(Collection<? extends Double> data)
          Creates a new instance of UnivariateGaussian from the given data
static StudentTDistribution.PDF learn(Collection<? extends Double> data, double defaultVariance)
          Creates a new instance of UnivariateGaussian from the given data
 
Methods inherited from class gov.sandia.cognition.util.AbstractCloneableSerializable
clone
 
Methods inherited from class java.lang.Object
equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
 
Methods inherited from interface gov.sandia.cognition.util.CloneableSerializable
clone
 

Field Detail

DEFAULT_VARIANCE

public static final double DEFAULT_VARIANCE
Typical value of a defaultVariance, 1.0E-5

See Also:
Constant Field Values
Constructor Detail

StudentTDistribution.MaximumLikelihoodEstimator

public StudentTDistribution.MaximumLikelihoodEstimator()
Default constructor


StudentTDistribution.MaximumLikelihoodEstimator

public StudentTDistribution.MaximumLikelihoodEstimator(double defaultVariance)
Creates a new instance of MaximumLikelihoodEstimator

Parameters:
defaultVariance - Amount to add to the variance to keep it from being 0.0
Method Detail

learn

public StudentTDistribution learn(Collection<? extends Double> data)
Creates a new instance of UnivariateGaussian from the given data

Specified by:
learn in interface BatchLearner<Collection<? extends Double>,StudentTDistribution>
Parameters:
data - Data to fit a UnivariateGaussian against
Returns:
Maximum likelihood estimate of the UnivariateGaussian that generated the data

learn

public static StudentTDistribution.PDF learn(Collection<? extends Double> data,
                                             double defaultVariance)
Creates a new instance of UnivariateGaussian from the given data

Parameters:
data - Data to fit a UnivariateGaussian against
defaultVariance - Amount to add to the variance to keep it from being 0.0
Returns:
Maximum likelihood estimate of the UnivariateGaussian that generated the data