Investigation of underlying distributional assumption in nested logit model using copula-based simulation and numerical approximation

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

This paper proposes a practical approach for drawing random samples that satisfy the underlying distributional assumptions in the nested logit model. This approach is a combined application of a series of statistical and numerical techniques including copula-based characterization of joint distributions, quasi-Monte Carlo integration, coordinate conversion, dichotomous approximation, and linear interpolation. With this approach, one may efficiently sample from and directly observe the distribution of common random components that result in correlation between random utilities in the nested logit model. Monte Carlo studies are conducted for two- and three-level nested logit models to evaluate the performance of random samples to recover model coefficients in nested logit models. Simulation experiments show that the relative differences are no more than 3% between mean values of estimators and their true values, indicating the proposed approach's high level of accuracy in reproducing random components in nested logit models.

Original languageEnglish (US)
Pages (from-to)36-43
Number of pages8
JournalTransportation Research Record
Issue number2254
DOIs
StatePublished - Dec 1 2011
Externally publishedYes

Fingerprint

Interpolation
Experiments

ASJC Scopus subject areas

  • Civil and Structural Engineering
  • Mechanical Engineering

Cite this

@article{d9dfce73ca91424aa7384eece8d15e0f,
title = "Investigation of underlying distributional assumption in nested logit model using copula-based simulation and numerical approximation",
abstract = "This paper proposes a practical approach for drawing random samples that satisfy the underlying distributional assumptions in the nested logit model. This approach is a combined application of a series of statistical and numerical techniques including copula-based characterization of joint distributions, quasi-Monte Carlo integration, coordinate conversion, dichotomous approximation, and linear interpolation. With this approach, one may efficiently sample from and directly observe the distribution of common random components that result in correlation between random utilities in the nested logit model. Monte Carlo studies are conducted for two- and three-level nested logit models to evaluate the performance of random samples to recover model coefficients in nested logit models. Simulation experiments show that the relative differences are no more than 3{\%} between mean values of estimators and their true values, indicating the proposed approach's high level of accuracy in reproducing random components in nested logit models.",
author = "Xin Ye",
year = "2011",
month = "12",
day = "1",
doi = "10.3141/2254-05",
language = "English (US)",
pages = "36--43",
journal = "Transportation Research Record",
issn = "0361-1981",
publisher = "US National Research Council",
number = "2254",

}

TY - JOUR

T1 - Investigation of underlying distributional assumption in nested logit model using copula-based simulation and numerical approximation

AU - Ye, Xin

PY - 2011/12/1

Y1 - 2011/12/1

N2 - This paper proposes a practical approach for drawing random samples that satisfy the underlying distributional assumptions in the nested logit model. This approach is a combined application of a series of statistical and numerical techniques including copula-based characterization of joint distributions, quasi-Monte Carlo integration, coordinate conversion, dichotomous approximation, and linear interpolation. With this approach, one may efficiently sample from and directly observe the distribution of common random components that result in correlation between random utilities in the nested logit model. Monte Carlo studies are conducted for two- and three-level nested logit models to evaluate the performance of random samples to recover model coefficients in nested logit models. Simulation experiments show that the relative differences are no more than 3% between mean values of estimators and their true values, indicating the proposed approach's high level of accuracy in reproducing random components in nested logit models.

AB - This paper proposes a practical approach for drawing random samples that satisfy the underlying distributional assumptions in the nested logit model. This approach is a combined application of a series of statistical and numerical techniques including copula-based characterization of joint distributions, quasi-Monte Carlo integration, coordinate conversion, dichotomous approximation, and linear interpolation. With this approach, one may efficiently sample from and directly observe the distribution of common random components that result in correlation between random utilities in the nested logit model. Monte Carlo studies are conducted for two- and three-level nested logit models to evaluate the performance of random samples to recover model coefficients in nested logit models. Simulation experiments show that the relative differences are no more than 3% between mean values of estimators and their true values, indicating the proposed approach's high level of accuracy in reproducing random components in nested logit models.

UR - http://www.scopus.com/inward/record.url?scp=84857849169&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84857849169&partnerID=8YFLogxK

U2 - 10.3141/2254-05

DO - 10.3141/2254-05

M3 - Article

AN - SCOPUS:84857849169

SP - 36

EP - 43

JO - Transportation Research Record

JF - Transportation Research Record

SN - 0361-1981

IS - 2254

ER -