negloglik

R2026b

Negative loglikelihood of probability distribution

Syntax

nll = negloglik(pd)

Description

nll = negloglik(pd) returns the value of the negative loglikelihood function for the data used to fit the probability distribution pd.

Examples

Negative Log Likelihood for a Fitted Distribution

Load the sample data.

load carsmall

Create a Weibull distribution object by fitting it to the mile per gallon (MPG) data.

pd = fitdist(MPG,'Weibull')

pd = 
  WeibullDistribution

  Weibull distribution
    A = 26.5079   [24.8333, 28.2954]
    B = 3.27193   [2.79441, 3.83104]

Compute the negative log likelihood for the fitted Weibull distribution.

wnll = negloglik(pd)

wnll = 
327.4942

Negative Loglikelihood for a Kernel Distribution

Load the sample data. Fit a kernel distribution to the miles per gallon (MPG) data.

load carsmall;
pd = fitdist(MPG,'Kernel')

pd = 
  KernelDistribution

    Kernel = normal
    Bandwidth = 4.11428
    Support = unbounded

Compute the negative loglikelihood.

nll = negloglik(pd)

nll = 
327.3139

Input Arguments

pd — Probability distribution

probability distribution object

Probability distribution, specified as one of the following probability distribution objects created by fitdist or Distribution Fitter:

• BetaDistribution

• BinomialDistribution

• BirnbaumSaundersDistribution

• BurrDistribution

• EmpiricalDistribution

• ExponentialDistribution

• ExtremeValueDistribution

• GammaDistribution

• GeneralizedExtremeValueDistribution

• GeneralizedParetoDistribution

• HalfNormalDistribution

• InverseGaussianDistribution

• KernelDistribution

• LogisticDistribution

• LoglogisticDistribution

• LognormalDistribution

• NakagamiDistribution

• NegativeBinomialDistribution

• NormalDistribution

• PoissonDistribution

• RayleighDistribution

• RicianDistribution

• ShiftedExponentialDistribution

• ShiftedGammaDistribution

• ShiftedWeibullDistribution

• StableDistribution

• tLocationScaleDistribution

• WeibullDistribution

Output Arguments

nll — Negative loglikelihood

numeric value

Negative loglikelihood value for the data used to fit the distribution, returned as a numeric value.

Extended Capabilities

GPU Arrays

Accelerate code by running on a graphics processing unit (GPU) using Parallel Computing Toolbox™.

Usage notes and limitations:

• The function supports GPU arrays except when the specified probability distribution is a shifted exponential, shifted gamma, or shifted Weibull distribution.

For more information, see Run MATLAB Functions on a GPU (Parallel Computing Toolbox).

Version History

Introduced in R2013a

R2026b: Create shifted exponential, gamma, and Weibull distribution objects

You can create three new shifted probability distribution objects by using makedist, or by fitting data with fitdist or the Distribution Fitter app:

• ShiftedExponentialDistribution — Two-parameter Exponential Distribution

• ShiftedGammaDistribution — Three-parameter Gamma Distribution

• ShiftedWeibullDistribution — Three-parameter Weibull Distribution

Each object includes a location parameter theta such that the distribution starts at x=theta instead of 0. You can use the objects with generic functions such as pdf, cdf, and random.