|
on Discrete Choice Models |
| By: | Oehlmann, Malte; Weller, Priska; Meyerhoff, Jürgen |
| Abstract: | In this paper we investigate the influence of choice task complexity on the propensity to choose the status quo (SQ) alternative in discrete choice experiments. Task complexity is characterized in terms of the design dimensionality systematically varying the number of choice sets, alternatives, attributes and levels as well as the level range using 16 split samples. Moreover, we use the number of level changes across alternatives and entropy to capture further complexity effects. First, we show that the frequency of choosing the SQ and the number of those respondents who always stay at the SQ varies across designs. Using a count data model and a binary logit we observe that both figures are particularly influenced by the number of alternatives. By interacting the alternative-specific constant of the SQ with our complexity measures in a conditional and random parameter logit, we then find that the probability to choose the SQ decreases with the number of alternatives and with designs having three attribute levels. The propensity to stay at the SQ, however, increases with higher values of entropy, more choice sets, and designs with a wider level range. Significant effects of socio-demographic characteristics on SQ choices are present in all our models. |
| JEL: | Q51 C35 D03 |
| Date: | 2014 |
| URL: | http://d.repec.org/n?u=RePEc:zbw:vfsc14:100616&r=dcm |
| By: | Yang , Cathy; Toubia, Olivier |
| Abstract: | It is becoming increasingly easier for researchers and practitioners to collect eye tracking data during online preference measurement tasks. The authors develop a dynamic discrete choice model of information search and choice under bounded rationality, that they calibrate using a combination of eye-tracking and choice data. Their model extends the directed cognition model of Gabaix et al. (2006) by capturing fatigue, proximity effects, and imperfect memory encoding and by estimating individual-level parameters and partworths within a likelihood-based, hierarchical Bayesian framework. The authors show that modeling eye movements as the outcome of forward-looking utility maximization improves out-of-sample predictions, enables researchers and practitioners to use shorter questionnaires, and allows better discrimination between attributes. |
| Keywords: | Preference Measurement; Incentive Compatibility; Eye Tracking; Dynamic Discrete Choice Models |
| JEL: | D83 M31 |
| Date: | 2014–10–14 |
| URL: | http://d.repec.org/n?u=RePEc:ebg:heccah:1072&r=dcm |
| By: | Phoebe Koundouri; Osiel Gonzalez Davila; Theologos Pantelidis |
| Abstract: | This study exploits the data from a full-preference ranking Choice Experiment (CE) designed to investigate how respondents evaluate a set of proposed improvements in the Asopos water catchment in Greece. These improvements are following the prescriptions of the European Union Water Framework Directive (2000). We first estimate a rank-ordered logistic regression based on the full set of choices by respondents to calculate the willingness to pay (WTP) of respondents for each one of the three attributes considered in the CE (that is, environmental conditions, impact on the local economy and changes in the potential uses of water). The model is initially estimated for the full sample and then re-estimated twice for two sub-samples: the one includes only the residents of Athens and the other includes only the residents of Asopos. Afterwards, we examine the effect of various demographic and socio-economic factors (such as income, gender, age, employment and education) on the estimates of our model in an attempt to reveal any differences among respondents with different characteristics. Thus, our analysis simultaneously provides a robustness check on previous findings in the literature and additional information about how various demographic and socio-economic characteristics affect the evaluation of the selected attributes. |
| Keywords: | Choice experiment, Full-preference ranking, Logistic regression, Asopos River Basin, Environmental degradation, Water quality and quantity; Random utility maximization; Logit probabilities; Water Framework Directive; Residency-specific use and non-use val |
| JEL: | Q25 Q51 Q53 |
| Date: | 2015–02–05 |
| URL: | http://d.repec.org/n?u=RePEc:aue:wpaper:1505&r=dcm |
| By: | Blasch, Julia (Department of Economics, Colgate University); Turner, Robert (Department of Economics, Colgate University) |
| Abstract: | Using a contingent choice survey of US citizens, we investigate the influence of environmental art on individual willingness to purchase voluntary carbon offsets. In a split-sample experiment, we compare the stated preferences of survey respondents in two different treatment groups to the preferences of a control group. One treatment group is shown photographs that illustrate the impacts of climate change; the other is shown animated images that illustrate wind speeds and patterns for extreme weather events. While individuals seeing the photographs show a higher willingness to purchase voluntary offset than the control group, respondents seeing the animated images seem less willing to buy offsets. This result remains stable when accounting for preference heterogeneity related to prior knowledge about climate change issues. We hypothesize that the differential impacts of the two kinds of artistic images are due to a combination of factors influencing individual choices: emotional affect, cognitive interest, and preferences for the prevention of specific climate change impacts as well as, more generally, internalized and social norms for the mitigation of climate change. |
| Keywords: | environmental art, climate change, carbon offsetting, knowledge, norms, discrete choice experiment |
| JEL: | Q5 Z1 |
| Date: | 2015–01–01 |
| URL: | http://d.repec.org/n?u=RePEc:cgt:wpaper:2015-01&r=dcm |
| By: | Phoebe Koundouri; Osiel Gonzalez Davila; Amerissa Giannouli; Nikolaos Kourogenis (Department of Banking and Financial Management, University of Piraeus.); Anastasios Xepapadeas; Peter A. Xepapadeas |
| Abstract: | Open Access initiatives, corresponding to unrestricted online access to peer-reviewed scholarly research are developed around the world and supported by research organizations and institutions. European Commission has launched in August 2008 the OpenAIRE (Open Access Infrastructure for Research in Europe) project, supporting Open Access in scientific information and research output. OpenAIRE connects institutional and thematic repositories, Open Access journals and Current Research Information Systems (CRIS), developing and promoting interoperability mechanisms for the efficient dissemination of scientific content, and monitoring uptake of funders� Open Access policies. In this paper we assess the economic sustainability of the OpenAIRE project. We use a Cost-Benefit Analysis framework in order to evaluate and compare the costs and benefits of OpenAIRE services, and provide explicit recommendations on the project�s economic efficiency and sustainable existence. OpenAIRE costs are explicitly documented in the market. OpenAIRE benefits, however, are not directly linked to specific market prices. As such these benefits can only be retrieved via the use of non-market valuation econometric methods. In this paper we apply a non-market valuation method based on the results of a �Choice Experiment�. Our approach is based on the elicitation of individual stakeholders� stated preferences and their aggregation over their relevant populations, yielding estimates of the Total Economic Value (TEV) generated by OpenAIRE. To this end, we identified, prioritized and mapped the relevant stakeholders based on their familiarity with Open Access and OpenAIRE, and their willingness to engage with similar initiatives. Then, we created two different questionnaires for researchers and non-researchers, using a full preference ranking approach. For the researchers we used budget reallocation from teaching/research budget and from infrastructure/library/administration budget. We estimate that the average OpenAIRE stakeholder�s Willingness to Pay (WTP) is 1763.13 �/institution/year for the basic services provided by OpenAIRE. OpenAIRE stakeholders prefer to have more interoperability between research platforms and research output, better access to scientific results and compliance to Open Access mandates. Applying a cost benefit analysis, we estimated that the net social benefits for the basic services for 15 years add up to 5,724,000 �, which are at least 5 times higher than costs� present value. Moreover, using Jone�s R&D based model, we estimated the potential R&D effect from research suggests even larger benefits in the long run with a Benefit/Cost ratio of 71.82, 95.75 and 115.58 for 50, 75 and 100 years, respectively. Risk analysis supports the robustness of the study�s results under different assumptions on future costs and benefits. Subscriptions based on the estimated WTP and cost, institutional subsidies and public awareness were our main recommendations for the sustainable operation of OpenAIRE. This study contributes to the literature on monetary valuation of the benefits and costs of Open Access to scientific knowledge. |
| Keywords: | Open access, OpenAIRE, research, economic valuation, choice experiment, rank-ordered logit, cost benefit analysis |
| Date: | 2015–02–05 |
| URL: | http://d.repec.org/n?u=RePEc:aue:wpaper:1503&r=dcm |
| By: | Klaus Nowotny |
| Abstract: | This paper investigates the economic, labor market and institutional factors that make regions and countries attractive for highly skilled migrants vis-`a-vis lowskill migrants. Based on micro-data for 11 EU countries, a discrete choice model estimated at the NUTS-2 level shows that location decisions are not only determined by factors related to earnings opportunities, distance, networks, common language and colonial relationships, but also by institutional factors such as migration policy, the income tax system, or labor market institutions; it also lends some support to the welfare magnet hypothesis: a higher unemployment replacement rate increases the attractiveness of a country. The empirical analysis however reveals only minor differences in the effects of institutions on location decisions by skill level, limiting the scope for policy makers to affect the skill composition of migration. |
| Keywords: | Highly skilled migration, regional location decisions, institutions, migration policy |
| JEL: | F22 R23 C35 |
| Date: | 2015–01 |
| URL: | http://d.repec.org/n?u=RePEc:feu:wfewop:y:2015:m:1:d:0:i:78&r=dcm |
| By: | Giovanni Bruno (University Commercial Luigi Bocconi, Milan); Orietta Dessy (Ca'Foscari University of Venice) |
| Abstract: | We extend the univariate results in Wooldridge (2005) to multivariate probit models, proving the following. 1) Average partial effects (APEs) based on joint probabilities are consistently estimated by conventional multivariate probit models under general forms of conditionally independent latent heterogeneity (LH) as long as the only constraints beyond normalization, if any, are within-equation homogenous restrictions. The normalization of choice is not neutral to consistency in models with cross-equation parameter restrictions beyond normalization, such as those implemented by Stata's asmprobit command or in the panel probit model: if the normalization is through an error covariance matrix in correlation form, consistency breaks down unless the LH components are truly homoskedastic. This is substantial because an error covariance matrix in correlation form is the only normalization permitted by Stata's biprobit and mvprobit commands or Limdep's BIVARIATE PROBIT and MPROBIT. Covariance restrictions beyond normalizations generally conflict with an arbitrary covariance matrix for the LH components. The multinomial probit model with i.i.d. errors, implemented by Stata's mprobit, is a case in point. 2) Conditional independence of the LH components is not generally sufficient for consistent estimation of APEs on conditional probabilities. Consistency is restored by maintaining an additional independence assumption. This holds true whether or not the response variables are used as regressors. 3) The dimensionality benefit observed by Mullahy (2011) in the estimation of partial effects extends to APEs. We exploit this feature in the design of a simple procedure estimating APEs, which is both faster and more accurate than simulation-based codes, such as Stata's mvprobit and cmp. To demonstrate the finite-sample implications of our results, we carry out extensive Monte Carlo experiments with bivariate and trivariate probit models. Finally, we apply our procedure in (3) to Italian survey data of immigrants in order to estimate the APEs of a trivariate probit model of ethnic identity formation and economic performance. |
| Date: | 2014–11–13 |
| URL: | http://d.repec.org/n?u=RePEc:boc:isug14:10&r=dcm |