Abstract
This chapter gives an overview of discrete choice analysis techniques. First we present a reflection about the meaning of the words ‘discrete’ and ‘choice’. Then we provide an overview of the sorts of choices in passenger and freight transport that have been treated as discrete choice problems. The next section presents the basic random utility theory, upon which most discrete choice models have been based. Different types of discrete choice models are then discussed: the workhorse of discrete choice modelling—the multinomial logit model (MNL), the nested logit and other Generalised Extreme Value (GEV) models, the probit model, the mixed logit and latent class models, ordered response models and aggregate logit models. Then we briefly discuss the estimation of discrete choice models and their application for demand forecasting and for policy simulation. The last section contains a summary and conclusions and a discussion on the future research directions.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
- 2.
Non-linear specifications of the utility function, such as functions with logarithmic or Box-Cox transformations (e.g. [28]), are also possible.
- 3.
Strictly speaking there is also a ‘scale’ parameter, which reflects the variance of the random component of utility and is used for normalising the model. It is called ‘scale’ parameter, because it scales the β parameters in (8.2a)–(8.2c); a higher random variance leads to lower estimated values of the β’ s.
- 4.
There are other reasons in the discrete choice literature for including the error terms.
- 5.
The name ‘logit’ comes from the fact that in the model presented above we are actually looking at differences between utility functions and the difference of two Gumbel distributions follows a logistic distribution. A ‘logistic’ or ‘logit’ model, that was invoked by assuming a logistic distribution for the error terms, had been around long before the above RUM-based model was first presented by especially McFadden (see [10]).
- 6.
The same logsum variable also has a meaning in welfare analysis: this logsum can be converted into money units (e.g. using a costs coefficient) and then the difference between the monetised logsum for a project alternative (with some infrastructure project) and the one for a reference case (without the project) can be used as the change in consumer surplus as a result of the project [15].
- 7.
Yet this is not the same model as the ordered response model, that is described in Sect. 8.8 of this chapter. OGEV is a generalisation of MNL, whereas ordered response models assume the same (non-RUM) response mechanism behind all ordered categories.
- 8.
An exception, in passenger transport, is the SILVESTER model for Stockholm [23].
- 9.
For the latter problem an alternative category of models is formed by the count data models, such as the Poisson regression model (see [7]).
- 10.
In practice it is often even difficult to obtain plausible transport time and costs coefficients when estimating on aggregate data. Prof. Moshe Ben-Akiva once suggested here to assume a value of time distribution to allow for heterogeneity between shipments.
- 11.
An alternative for the difference form is the ‘ratio form’ where the right-hand side has Pi/Pj and xiw/xjw, which has the disadvantage that the choice of the base mode (in the denominator of the dependent variable) affects the estimation results and the elasticities from the model. The difference form does not have this disadvantage.
References
Ben-Akiva M, Bierlaire M (1999) Discrete choice methods and their applications to short-term travel decisions. In: Hall R (ed) Handbook of transportation science. Kluwer, Amsterdam
Ben-Akiva M, Lerman SR (1985) Discrete choice analysis, theory and application to travel demand. MIT Press, Cambridge, Massachusetts
Beuthe M, Jourquin J, Geerts JF, Koul à Ndjang’Ha C (2001) Freight transport demand elasticities: a geographic multimodal transportation network analysis. Transp Res 37:253–266
Bhat CR (2008) The multiple discrete-continuous extreme value (MDCEV) model: Role of utility function parameters, identification considerations, and model extensions. Transp Res B 42(3):274–303
Bhat CR (2008b) Duration modelling. In: Hensher DA, Button KJ (eds) Handbooks in transport 1. Elsevier, Amsterdam
Bhat CR, Pulugurta V (1998) A comparison of two alternative behavioural choice mechanisms for household car ownership decisions. Transp Res B 32(1):61–75
Cameron AC, Trivedi PK (1998) Regression analysis of count data. Cambridge University Press, Cambridge
Chorus CG (2010) A new model of random regret minimization. Eur J Trans Infrastruct Res 10(2):181–196
Chorus CG, de Jong GC (2011) Modeling experienced accessibility: an approximation. J Transp Geogr 19:1155–1162
Cramer JS (1990) The logit model, an introduction for economists. Edward Arnold, London
Daly A, Bierlaire M (2006) A general and operational representation of GEV models. Transp Res B 40:285–305
Daly AJ, Zachary S (1978) Improved multiple choice models. In: Hensher DA, Dalvi MQ (eds) Determinants of travel choice. Saxon House, Farnborough
de Jong GC (1990) An indirect utility model of car ownership and private car use. Eur Econ Rev 34:971–985
de Jong GC (1996) A disaggregate model system of vehicle holding duration, type choice and use. Transp Res B 30(4):263–276
de Jong GC, Daly AJ, Pieters M, van der Hoorn AIJM (2007) The logsum as an evaluation measure: review of the literature and new results. Transp Res A 41:874–889
Geweke J, Keane M, Runkle D (1994) Alternative computational approaches to inference in the multinomial probit model. Rev Econ Stat 76:609–632
Greene WH, Hensher DA (2012) Revealing additional dimensions of preference heterogeneity in a latent class mixed multinomial logit model. Appl Econ 1–6
Hess S, Fowler M, Adler T, Bahreinian A (2009) A joint model for vehicle type and fuel type. Paper presented at the European transport conference, Noordwijkerhout, The Netherlands
Hess S, Ben-Akiva M, Gopinath D, Walker J (2011) Advantages of latent class over continuous mixture of logit models. Working paper, ITS, University of Leeds, Leeds
Holguín-Veras J, Xu N, de Jong GC, Maurer H (2011) An experimental economics investigation of shipper-carrier interactions on the choice of mode and shipment size in freight transport. Netw Spat Econ 11:509–532
Kahneman D, Tversky A (1979) Prospect theory: an analysis of decision under risk. Econometrica 47:263–291
Kahneman D, Tversky A (1992) Advances in prospect theory: cumulative representation of uncertainty. J Risk Uncertainty 5(4):297–323
Kristoffersson I (2011) Impacts of time-varying cordon pricing: validation and application of mesoscopic model for Stockholm. Transp Policy. doi:10.1016/j.tranpol.2011.06.006
McFadden D (1974) Conditional logit analysis of qualitative choice-behaviour. In: Zarembka P (ed) Frontiers in Econometrics. Academic Press, New York
McFadden D (1978) Modelling the choice of residential location. In: Karlqvist A, Lundqvist L, Snickars F, Weibull J (eds) Spatial interaction theory and residential location. North-Holland, Amsterdam
McFadden D (1981) Econometric models of probabilistic choice. In: Manski C, McFadden D (eds) Structural analysis of discrete data: with econometric applications. The MIT Press, Cambridge, Massachusetts
Miller EJ (2003) Microsimulation. In: Goulias KG (ed) Transportation systems planning: methods and applications. CRC Press, Boca Raton
Picard G, Gaudry P (1998) Exploration of a Box-Cox logit model of intercity freight mode choice. Transp Res E 34(1):1–12
Puckett SM, Hensher DA (2006) Modelling interdependent behaviour as a sequentially administrated stated choice experiment: analysis of freight distribution chains. Paper presented at the International Association of travel behaviour research conference, Kyoto
Rust J (1987) Optimal replacement of GMC bus engines: an empirical model of Harold Zurcher. Econometrica 55(5):999–1033
Significance (2009) Effect op autobezit van omzetting van de BPM in de Kilometerprijs. Report for the Environmental Assessment Agency, Significance, The Hague
Significance and Panteia (2012) Aanpassingen specificaties BasGoed V2.0. Report for DVS Rijkswaterstaat, Significance and Panteia, The Hague/Zoetermeer
Small K (1987) A discrete choice model for ordered alternatives. Econometrica 55(2):409–424
Tavasszy L, de Jong GC (2014) Modelling freight transport. Elsevier insights series. Elsevier, Amsterdam
Train K (1986) Qualitative choice analysis: theory, econometrics, and an application to automobile demand. MIT Press, Cambridge, Massachusetts
Train K (2003) Discrete choice methods with simulation. Cambridge University Press, Cambridge
van de Kaa EJ (2008) Extended prospect theory, findings on choice behaviour from economics and the behavioural sciences and their relevance for travel behaviour. Ph.D. thesis, Delft University of Technology, Delft
Vovsha P, Petersen E, Donnelly R (2003) Explicit modelling of joint travel by household members: statistical evidence and applied approach. Transp Res Rec 1831:1–10
Wen CH, Koppelman FS (2001) The generalised nested logit model. Transp Res B 35(7):627–641
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer International Publishing Switzerland
About this chapter
Cite this chapter
de Jong, G., Kroes, E. (2014). Discrete Choice Analysis. In: Lami, I. (eds) Analytical Decision-Making Methods for Evaluating Sustainable Transport in European Corridors. SxI - Springer for Innovation / SxI - Springer per l'Innovazione, vol 11. Springer, Cham. https://doi.org/10.1007/978-3-319-04786-7_8
Download citation
DOI: https://doi.org/10.1007/978-3-319-04786-7_8
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-04785-0
Online ISBN: 978-3-319-04786-7
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)