It is an enduring question how to combine revealed preference (RP) and stated preference (SP) data to analyze individual choices. While the nested logit (NL) model is the classical way to address the question, this study presents multitask learning deep neural networks (MTLDNNs) as an alternative framework, and discusses its theoretical foundation, empirical performance, and behavioral intuition. We

Estimating installed-base effects for product adoption in the presence of unobserved heterogeneity is challenging since the typical solution of including fixed effects leads to inconsistent estimates in models with installed base. Narayanan and Nair (2013) highlights this problem and proposes a bias correction method as a solution to the problem. This research note proposes an alternative solution:

Machine Learning (ML) approaches are increasingly being investigated as an alternative to Random Utility Models (RUMs) for modelling passenger mode choice. These approaches have the potential to provide valuable insights into choice modelling research questions. However, the research and the methodologies used are fragmented. Whilst systematic reviews on RUMs for mode choice prediction have long existed

This paper tests for the existence of loss aversion, reference dependence and diminishing sensitivity in a discrete choice experiment (DCE). A status quo alternative is introduced in a DCE of nurses' job choices and modelled as an individual-specific third alternative representing the respondent's current job. This provides a feasible method for including a status quo, which changes the reference point

In a referendum in March 2016, the British public voted by a margin of 52 per cent to 48 per cent to leave the European Union. Unfortunately, the referendum question provides little information on the sort of relationship Britons desire with the EU. The purpose of this study is to use stated choice experiments (CEs) to understand and quantify what aspects of the relationship between Britain and the

Wildlife is widely becoming an important vehicle for rural development in most third-world countries across the globe. With wildlife, as with other conservation and development policies, policymakers are usually not informed about the needs and wants of poor rural households and roll out programmes that are not tailor made to suit their desires, which often results in policy failure. We use a survey-based

Quantum probability, first developed in theoretical physics, has recently been successfully used in cognitive psychology to model data from experiments that previously resisted effective modelling by classical methods. This has led to the development of choice models based on quantum probability, which have greater flexibility than standard models due to the implementation of complex numbers through

Researchers using discrete choice experiments (DCE) are often faced with the difficult decision of selecting which are the key attributes that must be included into their analysis. Previous literature on methods for attribute selection is not particularly well documented, frequently leaving researchers with a wide choice of attributes that could lead to complex choice tasks. Moreover, selecting attributes

This paper i) compares parametric and semi-parametric representations of unobserved heterogeneity in hierarchical Bayesian multinomial logit models and ii) applies these methods to infer distributions of willingness to pay for features of shared automated vehicle (SAV) services. Specifically, we compare the multivariate normal, the finite mixture of normals and the Dirichlet process mixture of normals

Individual-specific posterior distributions are an attractive tool for disentangling the tastes for each person in the sample. However, there exists some risks and certain limitations regarding their use. This study reviews and summarizes the theoretical literature about the individual-specific posterior distributions derived from the Mixed Logit model, focusing on their properties, limitations and

The proneness of stated preference methods to different biases has increased the popularity of incentivized choice experiments (ICE). ICEs, however, are not free from challenges. One such challenge when using ICEs in market valuation is that some product alternatives may not be available to incentivize experiments. In this context, withholding information about product availability could be considered

The aim of the study is to examine the potential effect of attitudes towards physical activity on bus utility in the context of a rural area where studies have shown that the level and opportunity for physical activity are generally low. The need to analyze attitudes remains a strong motivation for the application of integrated choice and latent variable models. As such, we integrated attitudes towards

Combining choice modelling with social network analysis, we show how the stochastic actor-oriented model for the co-evolution of networks and behavior (SAOM) can be used as a powerful statistical framework to empirically analyze network-related choices. We discuss the underlying assumptions of SAOMs and show that they can be interpreted to represent a random utility maximization model (RUM). Network-related

This paper presents an agent-based model of team reasoning in a social dilemma game. Starting from the conundrum of empirically high levels of cooperation in dilemma games, contradicting traditional utility maximisation assumptions of game theory, Bacharach (1999, 2006) developed a theory of team reasoning. The idea behind team reasoning is that agents do not try to maximise their own utility but make

Images, videos, and virtual reality have been widely used in the literature to portray complex attributes to survey respondents. It is reasonable to expect immersive videos will be increasingly used in the future due to their decreasing costs and potentially more accurate representation of reality. Nevertheless, the literature has not sufficiently tested their ecological validity, which can be defined

Health economists often use discrete choice experiments (DCEs) to predict behavior, as actual market data is often unavailable. Manski (1990) argues that due to the incompleteness of the hypothetical scenarios used in DCEs, substantial uncertainty surrounds stated choice. Uncertainty can be decomposed into “resolvable” and “unresolvable”; the former is expected to become resolved in actual choice,

In linear regressions, the ecological fallacy—the erroneous belief that aggregate-level coefficients coincide with individual-level coefficients—arises when individual outcomes depend on the group environment. This paper suggest that such “group effects” determine also the circumstances under which the ecological fallacy vanishes from basic discrete-choice models. In particular, when controlling for

Endogeneity arises in discrete choice models due to several factors and results in inconsistent estimates of the model parameters. In adaptive choice contexts such as choice-based recommender systems and adaptive stated preferences (ASP) surveys, endogeneity is expected because the attributes presented to an individual in a specific menu (or choice situation) depend on the previous choices of the same

This paper proposes the use of Most Typical (MT) and Most Ideal (MI) levels when an adaptive choice-based conjoint (ACBC) survey can only obtain a small sample size n from a small population size N. This situation arises when expert decision makers are surveyed from among important small populations such as executives of large companies or political leaders, for which the expert decision maker assumption

Logit mixture models have gained increasing interest among researchers and practitioners because of their ability to capture unobserved taste heterogeneity. Becker et al. (2018) proposed a Hierarchical Bayes (HB) estimator for logit mixtures with inter- and intra-consumer heterogeneity (defined as taste variations among different individuals and among different choices made by the same individual respectively)