Multinomial Choice Modeling with NLOGIT: Model Estimation

NLOGIT supports a wide variety of specifications for discrete choice modeling.

Multinomial Logit - Many Specifications
Random Effects MNL

  • Multinomial logit specification
  • Scaled multinomial logit
  • Random regret
  • Nonlinear utility functions
  • Individual specific choice sets
  • Random effects and random parameters
  • Fixed effects
  • Choice specific attributes and interactions of characteristics with constants
  • Attribute nonattendance
  • Partial effects and elasticities
  • Test procedures for IIA
  • Restricted choice sets
  • Estimation using revealed preference, sets of ranks or best/worst
  • Merge stated and revealed preference data sets

Nested Logit
Generalized Nested Logit

  • Up to four levels in nested logit models
  • Constrained inclusive value parameters
  • Partial effects decomposed at the levels in the tree
  • Save utilities, inclusive values, probabilities
  • FIML or two step estimation
  • Random utility specifications to constrain the model

Multinomial Probit

  • Up to 20 choices
  • GHK simulator
  • Unrestricted or restricted correlation matrix
  • IIA test
  • Heteroscedasticity and covariance heterogeneity
  • Panel data - multinomial, multiperiod probit

Mixed (Random Parameters) Logit

  • Up to 100 random parameters
  • Maximum simulated likelihood estimation
  • Pseudorandom draws or Halton sequences
  • Mixture of random and nonrandom parameters
  • Panel data structures - random effects
  • Stated choice data
  • Freely correlated random parameters
  • Unrestricted mixture of normal, lognormal, triangular, uniform and other distributions
  • Restrictions on means and/or variances of random parameters
  • Individual heterogeneity in means and variances of random parameters
  • Individual specific parameter estimates
  • Error components logit allows choice specific random effects
  • Generalized mixed logit
  • Latent class mixed logit
  • Estimation in willingness to pay space
  • Nonlinear utility functions

Heteroscedastic Extreme Value

  • Choice specific variances in MNL model
  • Equality restrictions and grouping choices
  • Homogeneity of variances test

Latent Class

  • Multinomial logit structure
  • MNL sub model for class probabilities
  • Panel data structure
  • Up to 30 latent classes
  • Random parameters latent class model
  • Cross class restrictions on parameters