ModelFamily Class

Represents a family of distributions for the dependent variable in a GeneralizedLinearModel.

Definition

Namespace: Extreme.Statistics
Assembly: Extreme.Numerics (in Extreme.Numerics.dll) Version: 8.1.23

public abstract class ModelFamily

Visual Basic

Public MustInherit Class ModelFamily

C++/CLI

public ref class ModelFamily abstract

[<AbstractClassAttribute>]
type ModelFamily = class end

Inheritance: Object → ModelFamily

Remarks

Use the ModelFamily class to specify the probability distribution of the errors in a generalized linear model regression analysis.

Set the ModelFamily property of the GeneralizedLinearModel to one of the predefined fields or methods:

Member	Description
Normal	The dependent variable has a normal distribution. This is the default.
Binomial	The dependent variable has a binomial distribution. This is used for logistic regression.
Gamma	The dependent variable has a gamma distribution.
InverseGaussian	The dependent variable has an inverse Gaussian distribution.
NegativeBinomial	The dependent variable has a negative binomial distribution. This is a method that requires the total number of cases as a parameter.
Poisson	The dependent variable has a Poisson distribution. This is used for count data in Poisson regression.

Most of the members of this class are used internally. Two members that may be of interest are the IsLinkFunctionCompatible(LinkFunction) method, which checks if a LinkFunction is compatible with the model family, and the CanonicalLinkFunction property, which returns the canonical link function for the model family.

Constructors

ModelFamily

Initializes a new instance of the ModelFamily class

Properties

CanonicalLinkFunction	Gets the canonical link function for the model family.
HasScaleParameter	Gets or sets whether the probability distribution contains a scale factor that may be estimated.

Methods

DerivativeOfVarianceFunction	Evaluates the derivative of the variance function for the model family.
Deviance	Returns the contribution to the deviance of a case.
Equals	Determines whether the specified object is equal to the current object. (Inherited from Object)
Finalize	Allows an object to try to free resources and perform other cleanup operations before it is reclaimed by garbage collection. (Inherited from Object)
GetHashCode	Serves as the default hash function. (Inherited from Object)
GetType	Gets the Type of the current instance. (Inherited from Object)
IsLinkFunctionCompatible	Returns whether a link function is compatible with the model family.
KernelLogLikelihood	Returns the part of the log-likelihood for an observation that depends on the model parameters.
MemberwiseClone	Creates a shallow copy of the current Object. (Inherited from Object)
NegativeBinomial	The negative binomial distribution.
RemainderLogLikelihood	Returns the part of the log-likelihood for an observation that does not depend on the model parameters.
SquaredDeviance	Gets the square of the deviance contribution of a case.
ToString	Returns a string that represents the current object. (Inherited from Object)
VarianceFunction	Evaluates the variance function for the model family.

Fields

Binomial	The binomial distribution.
Gamma	The gamma distribution.
InverseGaussian	The inverse Gaussian or inverse normal distribution.
Normal	The Normal distribution. This is the default.
Poisson	The Poisson distribution.