Towhidul Islam

Designing Discrete Choice Experiments: Do Optimal Designs Come at a Price?

In discrete choice experiments, design decisions are crucial for determining data quality and costs. While high statistical efficiency designs are desirable, they may come at a price if they increase the cognitive burden for respondents. We address this problem by designing 44 experiments that systematically vary numbers of attributes and attribute level differences. Our results for two product categories suggest that respondents systematically are less consistent in answering choice questions as statistical efficiency increases. This relationship holds regardless of the number of attributes and is statistically significant even if one accommodates preference heterogeneity. Implications for practice and future research are discussed.

Modeling the choices of individual decision-makers by combining efficient choice experiment designs with extra preference information

Abstract We show how to combine statistically efficient ways to design discrete choice experiments based on random utility theory with new ways of collecting additional information that can be used to expand the amount of available choice information for modeling the choices of individual decision makers. Here we limit ourselves to problems involving generic choice options and linear and additive indirect utility functions, but the approach potentially can be extended to include choice problems with non-additive utility functions and non-generic/labeled options/attributes. The paper provides several simulated examples, a small empirical example to demonstrate proof of concept, and a larger empirical example based on many experimental conditions and large samples that demonstrates that the individual models capture virtually all the variance in aggregate first choices traditionally modeled in discrete choice experiments.

Forecasting the Diffusion of Innovations: Implications for Time-Series Extrapolation

The selection of an S-shaped trend model is a common step in attempts to model and forecast the diffusion of innovations. From the innovation-diffusion literature on model selection, forecasting, and the uncertainties associated with forecasts, we derive four principles.

Modelling multinational telecommunications demand with limited data

Abstract Forecasting the diffusion of innovations in the telecommunications sector is a constantly recurring problem for national providers. The problem is characterised by short data series making the estimation of model parameters unreliable. However, the same innovation will be diffusing simultaneously in other national markets, although with a different start date. The use of this cross-sectional data in constructing innovation diffusion models is investigated here. Four models for pooling the cross-sectional data are described and two diffusion models are discussed although only one, the Gompertz model is used throughout. Three innovation data sets are used in the evaluation of the models: digital cellular telephones, ISDN connections and fax connections. The pooled diffusion forecasts proved to be more accurate in several comparisons relative to a naive benchmark and to individual forecasts when available.

Retirement Investor Risk Tolerance in Tranquil and Crisis Periods: Experimental Survey Evidence

We conduct a choice experiment to investigate the impact of the financial crisis of 2008 on retirement saver investment choice and risk aversion. Analysis of estimated individual risk parameters shows a small increase in mean risk aversion between the relatively tranquil period of early 2007 and the crisis conditions of late 2008. Investment preferences of survey respondents, estimated using the scale-adjusted version of a latent class choice model, also change during the crisis. We identify age and income as important determinants of preference classes in both surveys and age is also identified as a key determinant of variability (scale). Young and low income individuals make choices that are more consistent with standard mean-variance analysis but older and higher income individuals react positively to both higher returns and increasing risk in returns. Overall we find a mild moderating of retirement investor risk tolerance in 2008.

Modelling diffusion and replacement

Abstract Replacement sales represent about 75% of total sales for many consumer durables, thus modelling this component well should lead to better overall forecasting. This paper surveys and evaluates forecasting models for total sales of durables which include both a diffusion component and a replacement component. The diffusion process, describing the behaviour of first time purchasers, is represented by a version of the Bass model and several different models of the replacement process are examined. The choice of replacement model is shown to have a major effect on forecasting performance. Two approaches are explored, model parameters are estimated with and without the use of prior estimates of expected service lifetime. Some of the two parameter replacement models are shown to offer forecasting performance superior to that of single parameter models, but their main disadvantage is only a 30–50% probability of successful parameter estimation. A robust approach to forecasting total sales is offered in conclusion.

Using copulas to model repeat purchase behaviour – An exploratory analysis via a case study

We develop a model of the evolution of inter-purchase times for a consumer-packaged product. After the introduction of the product, a consumer waits to make the initial purchase, then either waits to repurchase or decides not to. A repurchasing consumer repeats this decision process. The components of the model are the repurchase probability and the density function of the time to repurchase at each stage of the purchasing cycle. Issues of interest are: the strength of the dependency between successive repurchase times; the number of repeat purchases before stability occurs; the effects of consumer characteristics on inter-purchase times. The model of individual purchasing behaviour can be transformed via simulation to produce sales time series for a given population. As an example, the model is estimated for a product using Australian panel data. The accuracy of the models prediction is compared with an existing model.

Modelling the dependence between the times to international adoption of two related technologies

Abstract The structure of the dependence between the times to adoption by a country of two related innovations, the fax and the cellular telephone, is modelled in two stages. The first stage is the choice of density function for the time to adoption. The second stage is describing the dependence relation. For the first stage, a Weibull density function is used with its scale factor adapted to account for the economic and technological environments in different countries. Environmental data are collected from several sources. Copulas are used to model the dependence relation, three single parameter copulas are considered, those due to Farlie-Gumbel-Morgenstern (FGM), Frank and Plackett. Their properties are described and a combined estimation of the copula and density function parameters carried out. The limitations of the FGM copula rule it out from further consideration. The other copulas coupled with the Weibull, using eight environmental variables, are shown to provide valuable insights into the effects of environmental variables on adoption times. Given that a country has adopted one technology, the model of the dependence relation is used to provide the conditional density of the time to adoption of the other technology.

Consumer preferences for electronic cigarettes: results from a discrete choice experiment

Introduction E-cigarettes present a formidable challenge to regulators given their variety and the rapidly evolving nicotine market. The current study sought to examine the influence of e-cigarette product characteristics on consumer perceptions and trial intentions among Canadians. Methods An online discrete choice experiment was conducted with 915 Canadians aged 16 years and older in November 2013. An online commercial panel was used to sample 3 distinct subpopulations: (1) non-smoking youth and young adults (n=279); (2) smoking youth and young adults (n=264) and (3) smoking adults (n=372). Participants completed a series of stated-preference tasks, in which they viewed choice sets with e-cigarette product images that featured different combinations of attributes: flavour, nicotine content, health warnings and price. For each choice set, participants were asked to select one of the products or indicate ‘none of the above’ with respect to the following outcomes: interest in trying, less harm and usefulness in quitting smoking. The attributes’ impact on consumer choice for each outcome was analysed using multinomial logit regression. Results Health warning was the most important attribute influencing participants’ intentions to try e-cigarettes (42%) and perceived efficacy as a quit aid (39%). Both flavour (36%) and health warnings (35%) significantly predicted perceptions of product harm. Conclusions The findings indicate that consumers make trade-offs with respect to e-cigarette product characteristics, and that these trade-offs vary across different subpopulations. Given that health warnings and flavour were weighted most important by consumers in this study, these may represent good targets for e-cigarette regulatory frameworks.

Conceptual Relations Between Expanded Rank Data and Models of the Unexpanded Rank Data

Louviere et al. (2008, J. of Choice Modelling, 1, 126–163) present two main empirical examples in which a respondent rank orders the options in various choice sets by repeated best, then worst, choice. They expand the ranking data to various “implied” choices in subsets and fit the expanded data in various ways; they do not present models of the original rank data, except in one case (that of the rank ordered logit). We build on that work by constructing models of the original rank data that are consistent with the “weights” implied by the data expansions. This results in two classes of models: the first includes the reversible ranking model and has useful “score” properties; the second includes the rank ordered logit model and has natural “process” interpretations. We summarize known and new results on relations between the two classes of models and present fits of the models to the data of a case study concerning micro-generation of electricity using solar panels – that is, where individual households generate electricity using a renewable energy technology.