Variance-component-based nested logit specifications: Improved formulation, and practical microsimulation of random disturbance terms

David S. Bunch, David M Rocke

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

The initial motivation leading to the results in this paper was a problem most choice modeling researchers may have not considered: how to simulate random disturbance terms from nested logit (NL) models. We develop an approach using results from Cardell (1997), who proved the existence of a probability distribution (C(λ)) that could be used to formulate NL models based on statistically independent variance components. These components can be interpreted as unobserved preference heterogeneity for the choice 'dimensions' used to define NL tree structures. Simulation aside, we consider this formulation to have other practical advantages for empirical work, but it does not appear to have penetrated the literature (possibly due to notational obstacles). We use notation from Daly (2001) to implement an equivalent representation, which also establishes mathematical equivalence between Cardell (1997) and other important results in the NL literature.

Original languageEnglish (US)
JournalJournal of Choice Modelling
DOIs
StateAccepted/In press - Oct 4 2015

Fingerprint

Microsimulation
Logit
Variance Components
Nested Models
Logit Model
Disturbance
Specification
Specifications
Formulation
Term
Probability distributions
Tree Structure
Notation
Probability Distribution
Equivalence
Model-based
Modeling
Nested logit
Nested logit model
Variance components

Keywords

  • Discrete choice modeling
  • Monte Carlo simulation
  • Nested logit
  • Random disturbance terms
  • Random utility maximization

ASJC Scopus subject areas

  • Modeling and Simulation
  • Statistics, Probability and Uncertainty

Cite this

@article{98bc91bc29994847ae50cd752445eec0,
title = "Variance-component-based nested logit specifications: Improved formulation, and practical microsimulation of random disturbance terms",
abstract = "The initial motivation leading to the results in this paper was a problem most choice modeling researchers may have not considered: how to simulate random disturbance terms from nested logit (NL) models. We develop an approach using results from Cardell (1997), who proved the existence of a probability distribution (C(λ)) that could be used to formulate NL models based on statistically independent variance components. These components can be interpreted as unobserved preference heterogeneity for the choice 'dimensions' used to define NL tree structures. Simulation aside, we consider this formulation to have other practical advantages for empirical work, but it does not appear to have penetrated the literature (possibly due to notational obstacles). We use notation from Daly (2001) to implement an equivalent representation, which also establishes mathematical equivalence between Cardell (1997) and other important results in the NL literature.",
keywords = "Discrete choice modeling, Monte Carlo simulation, Nested logit, Random disturbance terms, Random utility maximization",
author = "Bunch, {David S.} and Rocke, {David M}",
year = "2015",
month = "10",
day = "4",
doi = "10.1016/j.jocm.2016.04.001",
language = "English (US)",
journal = "Journal of Choice Modelling",
issn = "1755-5345",
publisher = "Elsevier BV",

}

TY - JOUR

T1 - Variance-component-based nested logit specifications

T2 - Improved formulation, and practical microsimulation of random disturbance terms

AU - Bunch, David S.

AU - Rocke, David M

PY - 2015/10/4

Y1 - 2015/10/4

N2 - The initial motivation leading to the results in this paper was a problem most choice modeling researchers may have not considered: how to simulate random disturbance terms from nested logit (NL) models. We develop an approach using results from Cardell (1997), who proved the existence of a probability distribution (C(λ)) that could be used to formulate NL models based on statistically independent variance components. These components can be interpreted as unobserved preference heterogeneity for the choice 'dimensions' used to define NL tree structures. Simulation aside, we consider this formulation to have other practical advantages for empirical work, but it does not appear to have penetrated the literature (possibly due to notational obstacles). We use notation from Daly (2001) to implement an equivalent representation, which also establishes mathematical equivalence between Cardell (1997) and other important results in the NL literature.

AB - The initial motivation leading to the results in this paper was a problem most choice modeling researchers may have not considered: how to simulate random disturbance terms from nested logit (NL) models. We develop an approach using results from Cardell (1997), who proved the existence of a probability distribution (C(λ)) that could be used to formulate NL models based on statistically independent variance components. These components can be interpreted as unobserved preference heterogeneity for the choice 'dimensions' used to define NL tree structures. Simulation aside, we consider this formulation to have other practical advantages for empirical work, but it does not appear to have penetrated the literature (possibly due to notational obstacles). We use notation from Daly (2001) to implement an equivalent representation, which also establishes mathematical equivalence between Cardell (1997) and other important results in the NL literature.

KW - Discrete choice modeling

KW - Monte Carlo simulation

KW - Nested logit

KW - Random disturbance terms

KW - Random utility maximization

UR - http://www.scopus.com/inward/record.url?scp=84971634576&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84971634576&partnerID=8YFLogxK

U2 - 10.1016/j.jocm.2016.04.001

DO - 10.1016/j.jocm.2016.04.001

M3 - Article

AN - SCOPUS:84971634576

JO - Journal of Choice Modelling

JF - Journal of Choice Modelling

SN - 1755-5345

ER -