Models and optimal designs for conjoint choice experiments including a no-choice option

In a classical conjoint choice experiment, respondents choose one profile from each choice set that has to be evaluated. However, in real life, the respondent does not always make a choice: often he/she does not prefer any of the options offered. Therefore, including a no-choice option in a choice set makes a conjoint choice experiment more realistic. In the literature, three different models are used to analyze the results of a conjoint choice experiment with a no-choice option: the no-choice multinomial logit model, the extended no-choice multinomial logit model, and the nested no-choice multinomial logit model. We develop optimal designs for the two most appealing of these models using the D-optimality criterion and the modified Fedorov algorithm and compare these optimal designs with a reference design, which is constructed while ignoring the no-choice option, in terms of estimation and prediction accuracy. We conclude that taking into account the no-choice option when designing a no-choice experiment only has a marginal effect on the estimation and prediction accuracy as long as the model used for estimation matches the data-generating model.     

Generalized Reverse Discrete Choice Models

Marketing practitioners and academics have shown a keen interest in the processes that drive consumers’ choices since the early work of Guadagni and Little (1982). Over the past decade or so, a number of alternative models have been proposed, implemented and analyzed. The common behavioral assumption that underlines these models of discrete choice is random utility maximization (RUM). The RUM assumption, in its simplest form, posits that a consumer with a finite set of brands to choose from chooses the brand that gives her the maximum amount of utility. An alternative approach would be to assume that consumers choose the alternative that offers them the least disutility. Our paper proposes and tests a broad class of generalized extreme value models based on this hypothesis. We model the decision process of the consumer the assumption random disutility minimization (RDM) and derive a new class of discrete choice models based on this assumption. Our findings reveal that there are significant theoretical and econometric differences between the discrete choice models derived from a RUM framework and the RDM framework proposed in this paper. On the theoretical front we find that the class of discrete choice models based on the assumption of disutility minimization is structurally different from the models in the literature. Further, the models in this class are available in closed form and exhibit the same flexibility as the GEV models proposed by McFadden (1978). In fact, the number of parameters are identical to and have the same interpretation as those obtained via RUM based GEV models. In addition to the theoretical differences we also uncover significant empirical insights. With the computing effort and time for both models being roughly the same this new set of models offers marketing academics and researchers a viable new tool with which to investigate discrete choice behavior.JEL Classification: C25, C35, M37, D12     

Current Issues in Discrete Choice Modeling

Until recently, computational constraints forced researchers in thediscrete choice area to limit themselves to very simple statistical models,such as the multinomial logit (MNL), in which choice probabilities could beevaluated quickly on a computer. But the MNL only makes sense as abehavioral model under very special circumstances. Recent advances incomputation make it possible to estimate richer behavioral models thatgenerate very complex choice probability expressions. This paper discusses anumber of possible avenues for future research in the discrete choice areain light of these developments.     

Modeling preference evolution in discrete choice models: A Bayesian state-space approach

We develop discrete choice models that account for parameter driven preference dynamics. Choice model parameters may change over time because of shifting market conditions or due to changes in attribute levels over time or because of consumer learning. In this paper we show how such preference evolution can be modeled using hierarchial Bayesian state space models of discrete choice. The main feature of our approach is that it allows for the simultaneous incorporation of multiple sources of preference and choice dynamics. We show how the state space approach can include state dependence, unobserved heterogeneity, and more importantly, temporal variability in preferences using a correlated sequence of population distributions. The proposed model is very general and nests commonly used choice models in the literature as special cases. We use Markov chain monte carlo methods for estimating model parameters and apply our methodology to a scanner data set containing household brand choices over an eight-year period. Our analysis indicates that preferences exhibit significant variation over the time-span of the data and that incorporating time-variation in parameters is crucial for appropriate inferences regarding the magnitude and evolution of choice elasticities. We also find that models that ignore time variation in parameters can yield misleading inferences about the impact of causal variables. This paper is based on the first author's doctoral dissertation.     

HEV Choice Models as a Search Engine for the Specification of Nested Logit Tree Structures

This paper illustrates how the heteroskedastic extreme value (HEV) model can serve as an effective search engine for identifying appropriate tree structures in hierarchical choice models, particularly the nested logit. This use of the HEV model exploits its ability to estimate unique variances, and hence unique scale parameters, for each alternative in a choice set. The analysis of variance can reveal tree structures that may not be obvious to analysts who tend to base their search strategy on intuitive tree structures. The reliance on behavioural intuition may miss out on the identification of the 'best' tree in an econometric sense. This note illustrates how the HEV model is used to search for the hierarchical domain in which a statistically preferred nested logit model is positioned.     

Habit Persistence in Time Series Models of Discrete Choice

We provide a framework for modelling habit persistence in choice that integrates vector auto-regressive and moving-average (VARMA) time-series models with random coefficient Multinomial Probit (MNP) models. We provide two classes of models. In the first we assume that the error in the utility function has a general VARMA structure, and in the second we assume that structure for the regression coefficients. We provide an interpretation of these two classes of models. As an illustration, we re-analyse the A.C. Nielsen Company 1986/1987 scanner panel data on ketchup purchases and compare our model with two alternative state dependence models.     

Independent decision making of adolescents regarding food choice

The adolescence period is the stage in the life cycle in which individuals begin to develop independent decision making related to their social environment including their dietary intake. The period is fundamental in the development of longer term eating habits that may be reflected in adulthood. This study aimed to investigate the decision‐making process of adolescents and their dietary behaviour in the achievement of a balanced diet and to determine the underlying factors that may affect these choices. To meet this aim, a three phase data collection was utilized. In phase 1, a baseline questionnaire (n = 239) measured factors affecting dietary behaviour and the barriers in the achievement of a balanced diet. Phase two employed and measured respondents’ (n = 235) independent decision making by rating nineteen menu choices which reflected all options of dietary guidelines. This was analyzed through conjoint analysis. In phase three a subsample (n = 55) from the first two phases completed a food map which determined underlying reasons for food decisions and choice behaviour. Results revealed that when adolescents are given free choice they tend to follow an unbalanced diet with some diets raising concerns for their current and future health. Fifty‐seven percent of respondents displayed serious concerns in their dietary choice of which 18% were identified as requiring immediate intervention. The most noticeable factors affecting dietary choice were the desire to eat foods high in fat and carbohydrates and low in fruit and vegetable consumption. Food‐mapping indicated prior knowledge and attitudes had no effect on the choices made but that that behaviour of parents in the preparation of meals at home and peers in the selection of snack choices had the most influence on the unbalanced diet.     

FDI location choice of Chinese multinationals in East and Southeast Asia: Traditional economic factors and institutional perspective

This paper investigates the factors determining foreign direct investment (FDI) location choices of Chinese multinational firms. We developed a conceptual framework that synthesizes traditional economic factors and institutional perspective. Then several hypotheses were developed in line with the framework and empirically tested using panel data of Chinese outward FDI to eight economies in East and Southeast Asia across a time period of thirteen years. Our findings suggest that institutional factors demonstrate a higher level of significance, complexity and diversity in determining FDI location choice in comparison with economic factors, while both types of factors influence the FDI location choice of Chinese multinational firms. We also found that the FDI location choices of Chinese firms have a dynamic nature, as statistical evidence indicates a heterogeneous response of Chinese FDI towards different economic groups and during different time periods.     

Modeling Methods for Discrete Choice Analysis

This paper introduces new forms, sampling and estimation approaches fordiscrete choice models. The new models include behavioral specifications oflatent class choice models, multinomial probit, hybrid logit, andnon-parametric methods. Recent contributions also include new specializedchoice based sample designs that permit greater efficiency in datacollection. Finally, the paper describes recent developments in the use ofsimulation methods for model estimation. These developments are designed toallow the applications of discrete choice models to a wider variety ofdiscrete choice problems.     

Hybrid Choice Models: Progress and Challenges

We discuss the development of predictive choice models that go beyond the random utility model in its narrowest formulation. Such approaches incorporate several elements of cognitive process that have been identified as important to the choice process, including strong dependence on history and context, perception formation, and latent constraints. A flexible and practical hybrid choice model is presented that integrates many types of discrete choice modeling methods, draws on different types of data, and allows for flexible disturbances and explicit modeling of latent psychological explanatory variables, heterogeneity, and latent segmentation. Both progress and challenges related to the development of the hybrid choice model are presented.     

Recent Progress on Endogeneity in Choice Modeling

We describe recent progress in several areas related to endogeneity, including: choice set formation and attention to attributes; interactions among decision-makers; respondents' strategic behavior in answering stated preference choices; models of multiple discrete/continuous choice; distributions of willingness-to-pay; and methods for handling traditionally endogenous explanatory variables.     

A simultaneous model of multiple-discrete choices of variety and quantity

In many categories, consumers purchase discrete quantities of multiple varieties. For example, when doing grocery shopping for cereals, consumers may purchase in each category three units of brand A, four of brand B, and one of brand C. These decisions are often influenced by nonlinear pricing strategies such as quantity discounts. Modeling such multiple-discrete choices is challenging, as they violate assumptions of standard choice models. In this research, the author introduces a computationally attractive choice model that simultaneously captures 1) variety, 2) discrete quantity, and 3) nonlinear pricing strategies, such as quantity discounts. The model assumes that consumers maximize variety of the choice outcome, while taking into account constraints on utilities of alternatives. Application of the proposed model to two datasets demonstrates the superior fit compared to several rival models. Counterfactual analyses demonstrate that the model is a valuable tool for assortment and pricing decisions.     

Location choices of multinational firms in Europe: The role of EU cohesion policy

Using data on 5509 foreign subsidiaries established in 50 regions of 8 EU countries over the period 1991-1999, we estimate a mixed logit model of the location choice of multinational firms in Europe. In particular, we focus on the role of EU Cohesion Policy in attracting foreign investors from both within and outside Europe. We find that, after controlling for the role of agglomeration economies as well as a number of other regional and country characteristics and allowing for a very flexible correlation pattern among choices, Structural and Cohesion funds allocated by the EU to laggard regions have indeed contributed to attracting multinationals. These policies as well as other determinants play a different role in the case of European investors as opposed to non-European ones.     

Discrete choice models of firms’ strategic decisions

This paper outlines the methods and applications related to the nascent area of empirical discrete games in marketing. Many key strategic decisions firms make involve discrete choices such as deciding the location of a new store, determining where in product space to position a product, or what options to offer in a service contract. These decisions are fairly complex and typically involve the consideration of a number of demand, cost, and competitive factors. What makes these discrete choices particularly interesting (and challenging to analyze) is that they are interrelated with the choices of other firms because firms take into account the actions of their competitors when making their own decisions. We describe the basic problem of dealing with interrelated discrete choices in a game-theoretic framework and present the various estimation methods available. A discussion of the existing applications and future research opportunities concludes the article.     

Does religion influence location choice in the hotel industry?

Relying on data from the Spanish hotel industry, this paper analyzes the role of informal institutional factors (IIF) in location choice. Earlier studies mostly use an aggregate level of cultural differences as informal institutional factors. We, however, go deeper into this concept and study the impact of two distinct but interrelated informal institutional factors, religion and language, on the location decisions of hotel chains. We resolve the overlapping problem between these two highly correlated IIF by means of a ‘layer’ measurement in Confirmatory Factor Analysis. Our results show that the higher the informal institutional differences (IID), the lower the presence of the hotels in the foreign country. While physical distance plays a key moderating role, formal institutional differences (FID) did not show any effect. We contribute by unravelling the role of language and religion in location choice in the internationalization process of service firms. Moreover, we test the moderating role of formal institutions in these decisions, thus combining the impact of formal and informal institutions on location choices in service firms.     

Random regret minimization for consumer choice modeling: Assessment of empirical evidence

This paper introduces to the field of marketing a regret-based discrete choice model for the analysis of multi-attribute consumer choices from multinomial choice sets. This random regret minimization (RRM) model, which has recently been introduced in the field of transport, forms a regret-based counterpart of the canonical random utility maximization (RUM) paradigm. This paper assesses empirical results based on 43 comparisons reported in peer-reviewed journal articles and book chapters, with the aim of finding out to what extent, when, and how RRM can form a viable addition to the consumer choice modeler's toolkit. The paper shows that RRM and hybrid RRM–RUM models outperform RUM counterparts in a majority of cases, in terms of model fit and predictive ability. Although differences in performance are quite small, the two paradigms often result in markedly different managerial implications due to considerable differences in, for example, market share forecasts.     

Transnational corporations’ localization strategies via retail attributes: Focus on Chinese market

We explored the localization strategy of the transnational corporations (TNCs) in China by employing retail store attributes within the fast-changing Chinese retail market to select stores. We applied the multinomial logit (MNL) model based on McFadden’s (1974) random utility theory to proceed with this study. A questionnaire survey was conducted in nine areas in China. Surveys targeted people who had used Tesco, Walmart, and RT-Mart. 950 questionnaires were distributed, and 909 valid samples were obtained. The analysis results indicated that RT-Mart and Walmart, top players in the Chinese distribution industry, showed higher customer satisfaction in all sectors than Tesco, and statistically, all variables were significant. Customers of RT-Mart and Walmart were more satisfied compared with Tesco's customers in all variables, such as product quality, price, location, convenience, atmosphere, and staff service. In comparison to RT-Mart, Walmart had lower prices, fewer branches and convenience facilities, and lesser brand awareness, but the model estimates indicated that customer satisfaction was high regarding product quality, atmosphere, and employee service. In this study, various behavior factors were reviewed in consideration of consumer selection of stores; this aspect is significant because we have added attributes pertaining to the brand in our study.     

A Cross-Validity Comparison of Conjoint Analysis and Choice Models at Different Levels of Aggregation

Several (ratings-based) conjoint analysis and experimental choice (choice-based conjoint) models are compared on their ability to predict both aggregate choice shares among the sample and individual choices in an availability validation task. While there was a weak relationship between validations at the individual and aggregate levels, several models stand out. In general, models capturing individual differences validated well at both the individual and aggregate level. The hierarchical Bayes choice and conjoint models validated particularly well.Among choice models, the hierarchical Bayes choice model had the highest aggregate and individual level-validations. It was followed by the hybrid and seven segment latent segment choice models. Overall, the highest validating ratings-based conjoint model was the hierarchical Bayes model. However, the seven segment latent segment conjoint model produced better aggregate choice share validations than any other conjoint model. These results indicate that validations can be improved either by using benefit segment models and/or merging different types of data to estimate more individualized models.In most cases, rescaling improved the ratings-based, but not the choice-based choice share validations. This suggests that one might adjust for differences between ratings and choice tasks before making choice share predictions.     

Discrete and Continuous Representations of Unobserved Heterogeneity in Choice Modeling

We attempt to provide insights into how heterogeneity has been and can be addressed in choice modeling. In doing so, we deal with three topics: Models of heterogeneity, Methods of estimation and Substantive issues. In describing models we focus on discrete versus continuous representations of heterogeneity. With respect to estimation we contrast Markov Chain Monte Carlo methods and (simulated) likelihood methods. The substantive issues discussed deal with empirical tests of heterogeneity assumptions, the formation of empirical generalisations, the confounding of heterogeneity with state dependence and consideration sets, and normative segmentation.     

Information Spillovers and the Choice of Export Destination: A Multinomial Logit Analysis of Spanish Young SMEs

Does having more firms around exporting to a particular destination improve the chances of exporting to that destination (e.g. through information spillovers)? We answer this question implementing a multinomial logit model of whether a firm exports to a particular country. To identify the source of information spillovers, we construct indicators of geographical concentration of exporters selling to a specific destination: within industry, multinationals and across industries. In our application with data for Spanish new small sized firms, only within-industry agglomeration of exporting domestic firms significantly affects the probability of small sized firms exporting to the same destination. The significance of localisation economies is robust to a barrage of controls including destination specific characteristics, gravitational factors (distance and level of development), firm heterogeneity (size) and regional differences.