MIXED MNL MODELS FOR DISCRETE RESPONSE

By Daniel McFadden, Kenneth Train. 2000

4,630 citations recorded by the source corpus.

Subfields: Economics and Econometrics, Marketing, Transportation

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Economics and Econometrics

In Marketing

In Transportation

Citations in the corpus

Counterfactual inference for consumer choice across many product categories (2019, methodology): “McFadden and Train (2000) show that any choice probabilities derived from random utility maximization can be approximated arbitrarily well by a appropriately chosen mixed logit model.”
Panel Data Discrete Choice Models of Consumer Demand (2013, methodology): “It has been studied extensively by Berry (1994), Berry et al (1995), Harris and Keane (1999), McFadden and Train (2000), Train (2003) and others.”
Climate change and the willingness to pay to reduce ecological and health risks from wastewater flooding in urban centers and the environment (2014, methodology): “In this study, we apply the mixed logit model (McFadden and Train, 2000; Train, 2003) to address some limitations of the standard conditional logit model (McFadden, 1974) such as homogenous preferences and the assumption of the independence of irrelevant alternatives (IIA) (Hausman and McFadden,…”
Rewarding rush-hour avoidance: A study of commuters' travel behavior. (2011, background): “MXL accommodates random taste variation, substitution patterns, and correlation in unobserved factors unrestricted over time (McFadden & Train, 2000) and can be derived under a variety of different specifications (Ben Akiva & Bolduc, 1996; Bhat, 1998).”
Location decisions of Japanese new manufacturing plants in China: a discrete-choice analysis (2006, background): “not rely on the IIA assumption for coefficient estimation ( McFadden and Train 2000; Train 2003).”
COMPARING ALTERNATIVE MODELS OF HETEROGENEITY IN CONSUMER CHOICE BEHAVIOR (2012, methodology): “These included the nested logit model (McFadden, 1978), the generalized extreme value (GEV) model (McFadden, 1978), the multinomial probit (MNP) model (Thurston, 1927), the mixed MNL or mixed logit (MIXL) model (Ben-Akiva and McFadden et al., 1997; McFadden and Train, 2000), the latent class (LC) mo”
Customers' Choice Among Retail Energy Suppliers: The Willingness-to-Pay for Service Attributes (2000, positive): “Importantly, McFadden and Train (2000) show that the assumption of an extreme value error term is benign: any choice model can be approximated to any degree of accuracy with a mixed logit, i.e., with a model whose final error term is iid extreme value.”
Likelihood approximation by numerical integration on sparse grids (2008, methodology): “…of diﬀerent general simulation methods (Bhat 2001, Hess, Train and Polak 2006). p Chiou and Walker (2007) discuss the identiﬁcation and estimation of the MML model. t McFadden and Train (2000) provide an introduction to this model and a discussion of its estimation by simulation methods.”
Consumer willingness to pay for traditional food products (2016, background): “Mixed logit models (McFadden and Train, 2000) can capture unobserved preference heterogeneity, whereas the MNL assumes homogeneous preferences for product attributes.”
Valuing biodiversity enhancement in New Zealand's planted forests: Socioeconomic and spatial determinants of willingness-to-pay (2014, background): “It is based on random utility maximisation, but does not suffer from a series of 316 restrictive behavioural assumptions (McFadden and Train, 2000).”
Product Positioning and Competition: The Role of Location in the Fast Food Industry (2007, background): “9 See Anderson, de Palma and Thisse (1992), p. 350. substitutes as random coefficients play in many papers (McFadden and Train (2000), Sudhir (2001), Chintagunta, Dube and Singh (2003)), or that the variance-covariance terms play in papers using probit demand choice (Chintagunta (2001)).”
Halton Sequences for Mixed Logit (2000, methodology): “As McFadden and Train (1997) point out, this formula correctly incorporates simulation noise, unlike the commonly used (and easier to calculate) G-1.”
Willingness to pay for environmental attributes of non-food agricultural products: a real choice experiment (2013, methodology): “Following Revelt and Train (1998), we assume the price coefficient to be fixed in the population.”
Latent class model or mixed logit model? A comparison by transport mode choice data (2009, methodology): “For the other literature on the MLM and LCM, see Revelt and Train (1998), Louviere et al. (2000), McFadden and Train (2000), Bhat and Gossen (2004), Bishop and Provencher (2004), Greene (2003), Hensher et al. (2005), Train (2003), Train and Sandor (2004), Shen (2006), etc. 3Greene and Hensher…”