|
on Discrete Choice Models |
| By: | Davis, Katrina J; Burton, Michael; Kragt, Marit E |
| Abstract: | Models to analyse discrete choice data that account for heterogeneity in error variance (scale) across respondents are increasingly common, e.g. heteroscedastic conditional logit or scale adjusted latent class models. In this paper we do not question the need to allow for scale heterogeneity. Rather, we examine the interpretation of results from these models. We provide five empirical examples using discrete choice experiments, analysed using conditional logit, heteroscedastic conditional logit, or scale adjusted latent class models. We show that analysts may incorrectly conclude that preferences are consistent across respondents even if they are not, or that classes of respondents may have (in)significant preferences for some or all attributes of the experiment, when they do not. We recommend that future studies employing scale heterogeneity models explicitly state scale factors for all samples, choice contexts, and/or latent scale classes, and report rescaled preference parameters for each of these groups. |
| Keywords: | Discrete choice experiments, Heteroscedastic conditional logit models, Scale adjusted latent class models, Interpretation of preferences, Best-practice reporting, Research Methods/ Statistical Methods, C10, C18, C51, Q51, |
| Date: | 2016–05–14 |
| URL: | http://d.repec.org/n?u=RePEc:ags:uwauwp:235373&r=dcm |
| By: | Hartmann, Monika; Cash, Sean B.; Yeh, Ching-Hua; Landwehr, Stefanie C.; McAlister, Anna R. |
| Abstract: | Background: Children’s dietary related diseases and their associated costs have expanded dramatically in many countries, making children’s food choice a policy issue of increasing relevance. As children spend a considerable amount of money on energy dense, nutrient‐poor (EDNP) food products a better understanding of the main drivers of children’s food purchase decisions is crucial to prompt this behavior towards a more health promoting diet. Objective: The objective of the study is to investigate the role of branding and price in motivating children to choose healthier snack options. Methods: The study investigates snack choices of children ages 8 to 11 based on a survey and a purchase experiment. The research took place in after‐school programs of selected schools in the Boston area. 118 children took part in the study. Products in the choice experiment differed on three attributes, namely, product type (chocolate chip cookie as less healthy, and apple slices and strawberry tube yogurt as more healthy snacks), brand (McDonald’s or generic), and price ($0.30, $0.50, or $0.70). Data was analyzed using aggregated logit models, random parameter logit model and latent class analysis. Results: The results show that children’s purchase deci¬sions are primarily determined by product type with most children showing a high and signifi¬cant pre¬ference for choco¬late chip cookies. Surprisingly, generic products are preferred over the McDonald’s products across the whole sample, though children stating that they like McDonald’s reveal this also in their purchase decision. Prices only prove significant after controlling for whether or not children obtain allowance. Conclusion: First, it is not simple brand awareness but a child’s liking of the brand what determines whether a brand is success¬ful in motivating a child to choose a product and potentially a healthier option. Second, the extent of chil¬dren’s experience with money influences their price responsiveness. To the extent that children who receive allowance are primarily the once buying food snacks, higher prices for EDNP snacks could be success¬ful in motivating children to choose the healthier option. |
| Keywords: | children’s food preference, discrete choice experiment, aggregated and random parameter logit models, Food Consumption/Nutrition/Food Safety, Marketing, |
| Date: | 2016 |
| URL: | http://d.repec.org/n?u=RePEc:ags:aaea16:235841&r=dcm |
| By: | Aguirregabiria, Victor; Magesan, Arvind |
| Abstract: | This paper extends the Euler Equation (EE) representation of dynamic decision problems to a general class of discrete choice models and shows that the advantages of this approach apply not only to the estimation of structural parameters but also to the computation of a solution and to the evaluation of counterfactual experiments. We use a choice probabilities representation of the discrete decision problem to derive marginal conditions of optimality with the same features as the standard EEs in continuous decision problems. These EEs imply a fixed point mapping in the space of conditional choice values, that we denote the Euler equation-value (EE-value) operator. We show that, in contrast to Euler equation operators in continuous decision models, this operator is a contraction. We present numerical examples that illustrate how solving the model by iterating in the EE-value mapping implies substantial computational savings relative to iterating in the Bellman equation (that requires a much larger number of iterations) or in the policy function (that involves a costly valuation step). We define a sample version of the EE-value operator and use it to construct a sequence of consistent estimators of the structural parameters, and to evaluate counterfactual experiments. The computational cost of evaluating this sample-based EE-value operator increases linearly with sample size, and provides an unbiased (in finite samples) and consistent estimator the counterfactual. As such there is no curse of dimensionality in the consistent estimation of the model and in the evaluation of counterfactual experiments. We illustrate the computational gains of our methods using several Monte Carlo experiments. |
| Keywords: | Dynamic programming discrete choice models; Euler equations; Policy iteration; Estimation; Approximation bias. |
| JEL: | C13 C35 C51 C61 |
| Date: | 2016–05 |
| URL: | http://d.repec.org/n?u=RePEc:cpr:ceprdp:11300&r=dcm |
| By: | Victor Aguirregabiria; Arvind Magesan |
| Abstract: | This paper extends the Euler Equation (EE) representation of dynamic decision problems to a general class of discrete choice models and shows that the advantages of this approach apply not only to the estimation of structural parameters but also to the computation of a solution and to the evaluation of counterfactual experiments. We use a choice probabilities representation of the discrete decision problem to derive marginal conditions of optimality with the same features as the standard EEs in continuous decision problems. These EEs imply a fixed point mapping in the space of conditional choice values, that we denote the Euler equation-value (EE-value) operator. We show that, in contrast to Euler equation operators in continuous decision models, this operator is a contraction. We present numerical examples that illustrate how solving the model by iterating in the EE-value mapping implies substantial computational savings relative to iterating in the Bellman equation (that requires a much larger number of iterations) or in the policy function (that involves a costly valuation step). We define a sample version of the EE-value operator and use it to construct a sequence of consistent estimators of the structural parameters, and to evaluate counterfactual experiments. The computational cost of evaluating this sample-based EE-value operator increases linearly with sample size, and provides an unbiased (in finite samples) and consistent estimator the counterfactual. As such there is no curse of dimensionality in the consistent estimation of the model and in the evaluation of counterfactual experiments. We illustrate the computational gains of our methods using several Monte Carlo experiments. |
| Keywords: | Dynamic programming discrete choice models; Euler equations; Policy iteration; Estimation; Approximation bias |
| JEL: | C13 C35 C51 C61 |
| Date: | 2016–05–24 |
| URL: | http://d.repec.org/n?u=RePEc:tor:tecipa:tecipa-562&r=dcm |
| By: | Klaiman, Kimberly; Ortega, David; Garnache, Cloe |
| Abstract: | An increase in the amount of packaging consumed in the U.S. has put pressure on companies to take responsibility for the entire life-cycle of their product. This study uses discrete choice experiments to assess consumer willingness to pay (WTP) for packaging materials and recyclability of a beverage product. A between subject design was used to analyze the effectiveness of indirect questioning in addressing issues of social desirability bias as well as the effects of information on consumer behavior. Consumer WTP for packaging material was highest for plastic packaging, followed by glass, carton and aluminum. Our empirical analysis reveals that indirect questioning results in WTP values for packaging recyclability that are 60% lower than those from direct questioning. We find that information from a video treatment had a significant and positive effect on consumer preferences and demand for packaging recyclability. Our results suggest that more scrutiny should be placed on studies that do not address social desirability bias when evaluating recycling behavior. |
| Keywords: | Packaging, recycling, choice experiment, preference heterogeneity, Consumer/Household Economics, Environmental Economics and Policy, Marketing, Resource /Energy Economics and Policy, |
| Date: | 2016 |
| URL: | http://d.repec.org/n?u=RePEc:ags:aaea16:235549&r=dcm |
| By: | Liao, Kenneth; Pouliot, Sébastien |
| Abstract: | This paper estimates the relative preferences of motorists for E10 and E85 from an intercept survey of motorists with flex-fuel vehicles at E85 fuel stations in Arkansas, California, Colorado, Iowa, and Oklahoma. The information collected includes prices observed at fuel stations, fuel choices by flex motorists, and responses to a series of opinion questions about ethanol and gasoline. We also proposed a hypothetical scenario to each motorist in which either the price of the fuel selected was increased or the price of the fuel not selected was decreased. We first estimate fuel preferences using the revealed preference data from the observed choices. We then use the stated preference data from the hypothetical price scenario to estimate preferences in empirical models that correct for endogeneity from unobservable demand shifters that carry over to the stated preference empirical model. We find that motorists significantly discount E85 compared to E10 even when accounting for the different energy content of the two fuels and that the distribution of willingness to pay for E85 does not vary significantly between regions, except for California where motorists are willing to pay significantly more for E85. |
| Keywords: | Ethanol, Gasoline, Renewable Fuel Standard, Willingness to pay, Intercept survey, Agricultural and Food Policy, Demand and Price Analysis, Environmental Economics and Policy, Institutional and Behavioral Economics, Marketing, Resource /Energy Economics and Policy, Q18, Q41, Q42, |
| Date: | 2016–05 |
| URL: | http://d.repec.org/n?u=RePEc:ags:aaea16:236107&r=dcm |
| By: | Shr, Yau-Huo; Ready, Richard |
| Keywords: | Discrete choice analysis, Nonmarket Valuation, Environmental Economics and Policy, C52, |
| Date: | 2016 |
| URL: | http://d.repec.org/n?u=RePEc:ags:aaea16:236174&r=dcm |
| By: | Britwum, Kofi; Yiannaka, Amalia |
| Abstract: | Research on recent pre-slaughter interventions in the beef industry, particularly vaccinations and direct-fed microbials, have proven their effectiveness in reducing E. coli contamination in beef. In spite of such evidences, adoption of these technologies have been minimal. This study determined consumer response and willingness to pay (WTP) for beef products from cattle vaccinated against E. coli and given direct-fed microbials, and evaluated multiple message frames and their persuasive impacts on WTP for the technologies. Respondents were grouped into six information treatments, and were exposed to gain-framed and loss-framed messages, a media food safety story, and combinations of the media story and the gain-framed and loss-framed messages. A survey which included a choice experiment targeted a representative, random sample of 1,879 residents across the U.S in July and August 2015. A random parameters logit model found that consumers preferred animal vaccines over direct-fed microbials, and preferred either intervention to none at all. Corroborating prospect theory’s loss aversion, the loss-framed message, and the combined loss-framed message with the media story were the most persuasive, inducing the highest WTP. These findings altogether present an optimistic outlook about consumers’ openness to these technologies, and are of interest to agents in the beef sector who influence the variety and presentation of consumption choices available to consumers. |
| Keywords: | direct-fed microbials, message framing, vaccines, willingness to pay., Consumer/Household Economics, Food Consumption/Nutrition/Food Safety, Institutional and Behavioral Economics, D11 D12 Q13, |
| Date: | 2016–05–25 |
| URL: | http://d.repec.org/n?u=RePEc:ags:aaea16:235884&r=dcm |
| By: | Liu, Qing; Zhou, Jiehong; Yan, Zhen |
| Abstract: | The present study attempts to separate the environmental motivation and healthy motivation of consumers' choice for pro-environmental products through choice experiment and latent class model. Moreover, the different motives behind pro-environmental purchase and its impact on heterogeneity of consumer preferences needs to be further examined. Data are collected by means of face-to-face interview in Zhejiang, Shanghai, Jiangsu and Guangdong of China with a total sample size of 477 consumers. The results reveal that the consumers who are willing to buy pro-environmental products are motivated by health benefits as well as environmental considerations due to the higher consciousness of food safety and eco-environment in China. However, the healthy attributes tend to prevail in consumers' motivations. Consumers who have stronger environmental motivation will show higher preferences for pro-environmental products. It is also found that these consumers have the following characteristics: higher perception and knowledge of pro-environmental products，lower income , convenient purchase and female. Our results have implications for the improvement of government's consumption policies and the precision marketing of producers to induce consumers' participation to buy pro-environmental products, which in turn do a great benefit to environmental-friendly production and sustainable environment. |
| Keywords: | pro-environmental purchase motives, consumer preference, choice experiment, latent class model, Agricultural and Food Policy, |
| Date: | 2016 |
| URL: | http://d.repec.org/n?u=RePEc:ags:aaea16:236261&r=dcm |
| By: | Agyekum, Michael; Donovan, Cynthia; Lupi, Frank |
| Abstract: | Cowpeas are a critical crop for food and income security for millions of people in Africa. Insect pests, however, are a major production constraint causing over 50% yield losses. Farmers indiscriminately apply expensive synthetic pesticides in spite of attendant health and environmental risks. Recognizing that current chemical control methods are costly and unsustainable for resource-poor farmers, an innovative integrated pest management strategy based on biocontrol agents is being explored for West Africa. Therefore, the objective of this study was to predict potential adoption of the innovative biocontrol strategy by determining important factors that explain farmers’ pest control decisions. To this end, we designed a choice experiment survey for 505 cowpea farmers in Benin. A conditional logit model was estimated. Survey findings indicate that farmers are aware of health hazards but persistently apply chemical pesticides out of necessity. Also, model results reveal that important factors driving pest control decisions include costs, labor, and potential yield losses. Finally, we show that social norms could be leveraged to enhance adoption of the biocontrol method. These findings have promising implications for on-going efforts to develop and disseminate sustainable biocontrol strategies that are eco-friendly and cost effective for smallholder farmers. |
| Keywords: | discrete choice experiment, IPM, biological control, cowpea, West Africa, Consumer/Household Economics, Crop Production/Industries, Food Security and Poverty, International Development, D12, D60, Q12, Q50, |
| Date: | 2016–05–25 |
| URL: | http://d.repec.org/n?u=RePEc:ags:aaea16:235993&r=dcm |
| By: | Caputo, Vincenzina; Nayga, M. Rodolfo Jr.; Sacchi, Giovanna; Scarpa, Riccardo |
| Abstract: | The fact that survey respondents do not attend to all the attributes presented in choice experiment surveys is fast becoming a key issue in CE studies. This study proposes a new method aimed at eliciting consumers’ stated attribute non-attendance (ANA) behavior at the levels of each attribute, and compares it with the commonly used stated ANA approach, where non-attendance behavior is captured at the attribute level (i.e., not levels of the attributes). Results generally indicate that respondents do indeed ignore some of the levels of an attribute, suggesting that capturing non-attendance behavior at the attribute level would be insufficient at accurately and totally capturing stated ANA behavior. This finding implies that future choice experiment studies should take ANA into account not only at the attribute level but also at the levels of an attribute, both when asking respondents their stated ANA behavior during the survey and also in CE model specifications. |
| Keywords: | stated attribute level non-attendance, serial attribute non-attendance, serial attribute level non-attendance, choice experiments, Consumer/Household Economics, Institutional and Behavioral Economics, |
| Date: | 2016–08–02 |
| URL: | http://d.repec.org/n?u=RePEc:ags:aaea16:236035&r=dcm |
| By: | Palma, Marco; Li, Yajuan; Vedenov, Dmitry; Bessler, David |
| Abstract: | The simulated choice probabilities in Mixed Logit are approximated numerically from a multidimensional integral with a mixing distribution from a multivariate density function of the random parameters. Theoretically the order in which the variables are estimated should not matter; however, due to the inherent simulation ‘noise’ the magnitude of the estimated coefficients differs depending on the arbitrarily selected order in which the random variables enter the estimation procedure. This problem is exacerbated with a low number of draws or if correlation among coefficients is allowed. If correlation among the random parameters is allowed the variable ordering effects arise from simulation noise and from the Cholesky factorization used to allow for correlation. Ignoring the potential ordering effects in simulated maximum likelihood estimation methods seriously compromises the ability for replicating the results and can inadvertently influence policy recommendations. The simulation noise is independent of the number of integrating dimensions for random draws, but it increases for Halton draws. Hence, better coverage is achieved with Halton draws for small integrating dimensions, but random draws provide better coverage for larger dimensions. |
| Keywords: | Cholesky, Halton draws, Mixing distribution, Random draws, Random Parameters, Simulation., Research Methods/ Statistical Methods, C25, C63, |
| Date: | 2016 |
| URL: | http://d.repec.org/n?u=RePEc:ags:aaea16:235990&r=dcm |
| By: | Reeling, Carson; Verdier, Valentin; Lupi, Frank |
| Abstract: | "Preference point" lotteries, under which the probability an applicant is drawn increase with their stock of preference points earned over time, are widely used to allocate access to many economically important natural resources (e.g., big game hunting opportunities). Lotteries form a natural choice experiment: by observing the opportunities for which an individual applies, the alternatives not chosen, the associated costs, the probability of winning a permit, etc., statistical inferences can be made about how individuals trade off site characteristics for cost. Knowledge of these trade-o ffs can then be used to estimate applicants' willingness to pay for site quality characteristics and site access. Two key features of recreationalists' choices under preference point lottery are (i) forward-looking behavior (since the odds of winning a permit depend on the accumulated stock of preference points) and (ii) equilibrium sorting (whereby individuals decide where to apply based on their expectations of others' choices and vice versa). We develop a novel revealed preference method for estimating individuals' willingness to pay for access to recreational opportunities allocated by preference point lottery that accounts for these two features. We apply our model to the case study of black bear hunting in Michigan. We estimate total willingness to pay for access to a small site to be nearly $200,000. |
| Keywords: | dynamic discrete choice model, equilibrium sorting, lottery, preference points, revealed preference, travel cost, Environmental Economics and Policy, Research Methods/ Statistical Methods, Resource /Energy Economics and Policy, C2, C5, D9, Q26, Q51, |
| Date: | 2016 |
| URL: | http://d.repec.org/n?u=RePEc:ags:aaea16:235673&r=dcm |
| By: | Vielma Delano, Sofia K.; Quagrainie, Kwamena K. |
| Abstract: | There has been increasing concern about the environmental impact of pharmaceutical accumulation in surface and groundwater over the last two decades. Several states in the U.S. have implemented medicine take-back programs to help mitigate problems associated with unused and unwanted pharmaceuticals in households. States bordering the Great Lakes have particularly been concerned about this issue. This study assessed the value of a pharmaceutical collection program based on the willingness to pay per prescription and willingness to pay per visit of current and potential participants of the program in the Great Lakes area. We found that 60 % of the population is willing to participate in a collection program, while 40 % is willing to pay to participate in the program. The estimated unconditional mean WTP from a Double Hurdle Model is $0.53 per prescription and $1.03 per visit; and with the conditional mean willingness to pay, $1.25 per prescription and $2.33 per visit. Total annual benefits for such programs given the number of households in the area are estimated to be $20.1 million when considering WTP per prescription, and $18.9 million for a single drop-off per year when considering WTP per visit. This information will help better inform program providers, researchers, policymakers, advocates and other interested parties. |
| Keywords: | Pharmaceutical disposal, Contingent Valuation, Willingness to Pay, Double Hurdle, Environmental Economics and Policy, Resource /Energy Economics and Policy, |
| Date: | 2016–05 |
| URL: | http://d.repec.org/n?u=RePEc:ags:aaea16:236031&r=dcm |
| By: | Qing, Ping; Hu, Wuyang |
| Abstract: | China has risen to be the largest red wine consumer in the world but related studies using disaggregated and consumer-based data are scarce. This article examines Chinese preferences and willingness to pay for different wine attributes through a recent national survey including a choice experiment. Results indicate that country of origin is still one of the most important attributes for wine. Taste of wine and organic production are also relevant to consumers. Wine vintage is not as important as expected. Key implications on Chinese domestic and imported wines are discussed under the context of recent profound structural changes in wine consumption induced by policy shifts. |
| Keywords: | China, wine, Agribusiness, Agricultural and Food Policy, Consumer/Household Economics, Food Consumption/Nutrition/Food Safety, Marketing, Q13, |
| Date: | 2016–05 |
| URL: | http://d.repec.org/n?u=RePEc:ags:aaea16:235477&r=dcm |