A hybrid choice model with nonlinear utility functions and bounded distributions for latent variables

Application to purchase intention decisions of electric cars

Jinhee Kim, Soora Rasouli, Harry J.P. Timmermans

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

This study suggests an extended hybrid choice model framework allowing nonlinear utility functions of choice alternatives including not only observable but also latent variables. The nonlinear utility function is developed based on the Box-Cox transformation. For satisfying the constraint of the Box-Cox transformation and being consistent with the bounded nature of the measurement of the indicators, we suggest using bounded distributions for the latent variables, such as the doubly censored normal distribution and Johnson's SB distribution. An empirical study is conducted using stated choice data about the intention to purchase electric cars. The empirical results show that the proposed framework can enhance the explanatory power of hybrid choice models compared to the model allowing only a linear specification.

Original languageEnglish
Title of host publicationProceedings of the 19th International Conference of Hong Kong Society for Transportation Studies, HKSTS 2014 - Transportation and Infrastructure
EditorsZ. Leng, Y. H. Wang
PublisherHong Kong Society for Transportation Studies Limited
Pages243-250
Number of pages8
ISBN (Electronic)9789881581433
Publication statusPublished - 2014 Jan 1
Event19th International Conference of Hong Kong Society for Transportation Studies: Transportation and Infrastructure, HKSTS 2014 - Hong Kong, China
Duration: 2014 Dec 132014 Dec 15

Publication series

NameProceedings of the 19th International Conference of Hong Kong Society for Transportation Studies, HKSTS 2014 - Transportation and Infrastructure

Other

Other19th International Conference of Hong Kong Society for Transportation Studies: Transportation and Infrastructure, HKSTS 2014
CountryChina
CityHong Kong
Period14/12/1314/12/15

Fingerprint

purchase
Railroad cars
Normal distribution
Specifications

All Science Journal Classification (ASJC) codes

  • Transportation

Cite this

Kim, J., Rasouli, S., & Timmermans, H. J. P. (2014). A hybrid choice model with nonlinear utility functions and bounded distributions for latent variables: Application to purchase intention decisions of electric cars. In Z. Leng, & Y. H. Wang (Eds.), Proceedings of the 19th International Conference of Hong Kong Society for Transportation Studies, HKSTS 2014 - Transportation and Infrastructure (pp. 243-250). (Proceedings of the 19th International Conference of Hong Kong Society for Transportation Studies, HKSTS 2014 - Transportation and Infrastructure). Hong Kong Society for Transportation Studies Limited.
Kim, Jinhee ; Rasouli, Soora ; Timmermans, Harry J.P. / A hybrid choice model with nonlinear utility functions and bounded distributions for latent variables : Application to purchase intention decisions of electric cars. Proceedings of the 19th International Conference of Hong Kong Society for Transportation Studies, HKSTS 2014 - Transportation and Infrastructure. editor / Z. Leng ; Y. H. Wang. Hong Kong Society for Transportation Studies Limited, 2014. pp. 243-250 (Proceedings of the 19th International Conference of Hong Kong Society for Transportation Studies, HKSTS 2014 - Transportation and Infrastructure).
@inproceedings{9399516f446745518a1fa080ca1ec569,
title = "A hybrid choice model with nonlinear utility functions and bounded distributions for latent variables: Application to purchase intention decisions of electric cars",
abstract = "This study suggests an extended hybrid choice model framework allowing nonlinear utility functions of choice alternatives including not only observable but also latent variables. The nonlinear utility function is developed based on the Box-Cox transformation. For satisfying the constraint of the Box-Cox transformation and being consistent with the bounded nature of the measurement of the indicators, we suggest using bounded distributions for the latent variables, such as the doubly censored normal distribution and Johnson's SB distribution. An empirical study is conducted using stated choice data about the intention to purchase electric cars. The empirical results show that the proposed framework can enhance the explanatory power of hybrid choice models compared to the model allowing only a linear specification.",
author = "Jinhee Kim and Soora Rasouli and Timmermans, {Harry J.P.}",
year = "2014",
month = "1",
day = "1",
language = "English",
series = "Proceedings of the 19th International Conference of Hong Kong Society for Transportation Studies, HKSTS 2014 - Transportation and Infrastructure",
publisher = "Hong Kong Society for Transportation Studies Limited",
pages = "243--250",
editor = "Z. Leng and Wang, {Y. H.}",
booktitle = "Proceedings of the 19th International Conference of Hong Kong Society for Transportation Studies, HKSTS 2014 - Transportation and Infrastructure",

}

Kim, J, Rasouli, S & Timmermans, HJP 2014, A hybrid choice model with nonlinear utility functions and bounded distributions for latent variables: Application to purchase intention decisions of electric cars. in Z Leng & YH Wang (eds), Proceedings of the 19th International Conference of Hong Kong Society for Transportation Studies, HKSTS 2014 - Transportation and Infrastructure. Proceedings of the 19th International Conference of Hong Kong Society for Transportation Studies, HKSTS 2014 - Transportation and Infrastructure, Hong Kong Society for Transportation Studies Limited, pp. 243-250, 19th International Conference of Hong Kong Society for Transportation Studies: Transportation and Infrastructure, HKSTS 2014, Hong Kong, China, 14/12/13.

A hybrid choice model with nonlinear utility functions and bounded distributions for latent variables : Application to purchase intention decisions of electric cars. / Kim, Jinhee; Rasouli, Soora; Timmermans, Harry J.P.

Proceedings of the 19th International Conference of Hong Kong Society for Transportation Studies, HKSTS 2014 - Transportation and Infrastructure. ed. / Z. Leng; Y. H. Wang. Hong Kong Society for Transportation Studies Limited, 2014. p. 243-250 (Proceedings of the 19th International Conference of Hong Kong Society for Transportation Studies, HKSTS 2014 - Transportation and Infrastructure).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - A hybrid choice model with nonlinear utility functions and bounded distributions for latent variables

T2 - Application to purchase intention decisions of electric cars

AU - Kim, Jinhee

AU - Rasouli, Soora

AU - Timmermans, Harry J.P.

PY - 2014/1/1

Y1 - 2014/1/1

N2 - This study suggests an extended hybrid choice model framework allowing nonlinear utility functions of choice alternatives including not only observable but also latent variables. The nonlinear utility function is developed based on the Box-Cox transformation. For satisfying the constraint of the Box-Cox transformation and being consistent with the bounded nature of the measurement of the indicators, we suggest using bounded distributions for the latent variables, such as the doubly censored normal distribution and Johnson's SB distribution. An empirical study is conducted using stated choice data about the intention to purchase electric cars. The empirical results show that the proposed framework can enhance the explanatory power of hybrid choice models compared to the model allowing only a linear specification.

AB - This study suggests an extended hybrid choice model framework allowing nonlinear utility functions of choice alternatives including not only observable but also latent variables. The nonlinear utility function is developed based on the Box-Cox transformation. For satisfying the constraint of the Box-Cox transformation and being consistent with the bounded nature of the measurement of the indicators, we suggest using bounded distributions for the latent variables, such as the doubly censored normal distribution and Johnson's SB distribution. An empirical study is conducted using stated choice data about the intention to purchase electric cars. The empirical results show that the proposed framework can enhance the explanatory power of hybrid choice models compared to the model allowing only a linear specification.

UR - http://www.scopus.com/inward/record.url?scp=84928029090&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84928029090&partnerID=8YFLogxK

M3 - Conference contribution

T3 - Proceedings of the 19th International Conference of Hong Kong Society for Transportation Studies, HKSTS 2014 - Transportation and Infrastructure

SP - 243

EP - 250

BT - Proceedings of the 19th International Conference of Hong Kong Society for Transportation Studies, HKSTS 2014 - Transportation and Infrastructure

A2 - Leng, Z.

A2 - Wang, Y. H.

PB - Hong Kong Society for Transportation Studies Limited

ER -

Kim J, Rasouli S, Timmermans HJP. A hybrid choice model with nonlinear utility functions and bounded distributions for latent variables: Application to purchase intention decisions of electric cars. In Leng Z, Wang YH, editors, Proceedings of the 19th International Conference of Hong Kong Society for Transportation Studies, HKSTS 2014 - Transportation and Infrastructure. Hong Kong Society for Transportation Studies Limited. 2014. p. 243-250. (Proceedings of the 19th International Conference of Hong Kong Society for Transportation Studies, HKSTS 2014 - Transportation and Infrastructure).