A practical method to test the validity of the standard Gumbel distribution in logit-based multinomial choice models of travel behavior

Xin Ye, Venu M. Garikapati, Daehyun You, Ram Pendyala

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

Most multinomial choice models (e.g., the multinomial logit model) adopted in practice assume an extreme-value Gumbel distribution for the random components (error terms) of utility functions. This distributional assumption offers a closed-form likelihood expression when the utility maximization principle is applied to model choice behaviors. As a result, model coefficients can be easily estimated using the standard maximum likelihood estimation method. However, maximum likelihood estimators are consistent and efficient only if distributional assumptions on the random error terms are valid. It is therefore critical to test the validity of underlying distributional assumptions on the error terms that form the basis of parameter estimation and policy evaluation. In this paper, a practical yet statistically rigorous method is proposed to test the validity of the distributional assumption on the random components of utility functions in both the multinomial logit (MNL) model and multiple discrete-continuous extreme value (MDCEV) model. Based on a semi-nonparametric approach, a closed-form likelihood function that nests the MNL or MDCEV model being tested is derived. The proposed method allows traditional likelihood ratio tests to be used to test violations of the standard Gumbel distribution assumption. Simulation experiments are conducted to demonstrate that the proposed test yields acceptable Type-I and Type-II error probabilities at commonly available sample sizes. The test is then applied to three real-world discrete and discrete-continuous choice models. For all three models, the proposed test rejects the validity of the standard Gumbel distribution in most utility functions, calling for the development of robust choice models that overcome adverse effects of violations of distributional assumptions on the error terms in random utility functions.

Original languageEnglish (US)
Pages (from-to)173-192
Number of pages20
JournalTransportation Research Part B: Methodological
Volume106
DOIs
StatePublished - Dec 1 2017

Fingerprint

travel behavior
Random errors
Maximum likelihood estimation
Parameter estimation
Maximum likelihood
simulation

Keywords

  • Discrete choice models
  • Multinomial logit model
  • Multiple discrete-continuous extreme value model
  • Test of validity of distributional assumption
  • Travel behavior models
  • Violations of distributional assumptions

ASJC Scopus subject areas

  • Transportation

Cite this

A practical method to test the validity of the standard Gumbel distribution in logit-based multinomial choice models of travel behavior. / Ye, Xin; Garikapati, Venu M.; You, Daehyun; Pendyala, Ram.

In: Transportation Research Part B: Methodological, Vol. 106, 01.12.2017, p. 173-192.

Research output: Contribution to journalArticle

@article{edf1a20381c7433caa7fa062e38b1f4f,
title = "A practical method to test the validity of the standard Gumbel distribution in logit-based multinomial choice models of travel behavior",
abstract = "Most multinomial choice models (e.g., the multinomial logit model) adopted in practice assume an extreme-value Gumbel distribution for the random components (error terms) of utility functions. This distributional assumption offers a closed-form likelihood expression when the utility maximization principle is applied to model choice behaviors. As a result, model coefficients can be easily estimated using the standard maximum likelihood estimation method. However, maximum likelihood estimators are consistent and efficient only if distributional assumptions on the random error terms are valid. It is therefore critical to test the validity of underlying distributional assumptions on the error terms that form the basis of parameter estimation and policy evaluation. In this paper, a practical yet statistically rigorous method is proposed to test the validity of the distributional assumption on the random components of utility functions in both the multinomial logit (MNL) model and multiple discrete-continuous extreme value (MDCEV) model. Based on a semi-nonparametric approach, a closed-form likelihood function that nests the MNL or MDCEV model being tested is derived. The proposed method allows traditional likelihood ratio tests to be used to test violations of the standard Gumbel distribution assumption. Simulation experiments are conducted to demonstrate that the proposed test yields acceptable Type-I and Type-II error probabilities at commonly available sample sizes. The test is then applied to three real-world discrete and discrete-continuous choice models. For all three models, the proposed test rejects the validity of the standard Gumbel distribution in most utility functions, calling for the development of robust choice models that overcome adverse effects of violations of distributional assumptions on the error terms in random utility functions.",
keywords = "Discrete choice models, Multinomial logit model, Multiple discrete-continuous extreme value model, Test of validity of distributional assumption, Travel behavior models, Violations of distributional assumptions",
author = "Xin Ye and Garikapati, {Venu M.} and Daehyun You and Ram Pendyala",
year = "2017",
month = "12",
day = "1",
doi = "10.1016/j.trb.2017.10.009",
language = "English (US)",
volume = "106",
pages = "173--192",
journal = "Transportation Research, Series B: Methodological",
issn = "0191-2615",
publisher = "Elsevier Limited",

}

TY - JOUR

T1 - A practical method to test the validity of the standard Gumbel distribution in logit-based multinomial choice models of travel behavior

AU - Ye, Xin

AU - Garikapati, Venu M.

AU - You, Daehyun

AU - Pendyala, Ram

PY - 2017/12/1

Y1 - 2017/12/1

N2 - Most multinomial choice models (e.g., the multinomial logit model) adopted in practice assume an extreme-value Gumbel distribution for the random components (error terms) of utility functions. This distributional assumption offers a closed-form likelihood expression when the utility maximization principle is applied to model choice behaviors. As a result, model coefficients can be easily estimated using the standard maximum likelihood estimation method. However, maximum likelihood estimators are consistent and efficient only if distributional assumptions on the random error terms are valid. It is therefore critical to test the validity of underlying distributional assumptions on the error terms that form the basis of parameter estimation and policy evaluation. In this paper, a practical yet statistically rigorous method is proposed to test the validity of the distributional assumption on the random components of utility functions in both the multinomial logit (MNL) model and multiple discrete-continuous extreme value (MDCEV) model. Based on a semi-nonparametric approach, a closed-form likelihood function that nests the MNL or MDCEV model being tested is derived. The proposed method allows traditional likelihood ratio tests to be used to test violations of the standard Gumbel distribution assumption. Simulation experiments are conducted to demonstrate that the proposed test yields acceptable Type-I and Type-II error probabilities at commonly available sample sizes. The test is then applied to three real-world discrete and discrete-continuous choice models. For all three models, the proposed test rejects the validity of the standard Gumbel distribution in most utility functions, calling for the development of robust choice models that overcome adverse effects of violations of distributional assumptions on the error terms in random utility functions.

AB - Most multinomial choice models (e.g., the multinomial logit model) adopted in practice assume an extreme-value Gumbel distribution for the random components (error terms) of utility functions. This distributional assumption offers a closed-form likelihood expression when the utility maximization principle is applied to model choice behaviors. As a result, model coefficients can be easily estimated using the standard maximum likelihood estimation method. However, maximum likelihood estimators are consistent and efficient only if distributional assumptions on the random error terms are valid. It is therefore critical to test the validity of underlying distributional assumptions on the error terms that form the basis of parameter estimation and policy evaluation. In this paper, a practical yet statistically rigorous method is proposed to test the validity of the distributional assumption on the random components of utility functions in both the multinomial logit (MNL) model and multiple discrete-continuous extreme value (MDCEV) model. Based on a semi-nonparametric approach, a closed-form likelihood function that nests the MNL or MDCEV model being tested is derived. The proposed method allows traditional likelihood ratio tests to be used to test violations of the standard Gumbel distribution assumption. Simulation experiments are conducted to demonstrate that the proposed test yields acceptable Type-I and Type-II error probabilities at commonly available sample sizes. The test is then applied to three real-world discrete and discrete-continuous choice models. For all three models, the proposed test rejects the validity of the standard Gumbel distribution in most utility functions, calling for the development of robust choice models that overcome adverse effects of violations of distributional assumptions on the error terms in random utility functions.

KW - Discrete choice models

KW - Multinomial logit model

KW - Multiple discrete-continuous extreme value model

KW - Test of validity of distributional assumption

KW - Travel behavior models

KW - Violations of distributional assumptions

UR - http://www.scopus.com/inward/record.url?scp=85034454572&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85034454572&partnerID=8YFLogxK

U2 - 10.1016/j.trb.2017.10.009

DO - 10.1016/j.trb.2017.10.009

M3 - Article

VL - 106

SP - 173

EP - 192

JO - Transportation Research, Series B: Methodological

JF - Transportation Research, Series B: Methodological

SN - 0191-2615

ER -