|
on Discrete Choice Models |
| By: | Paul Chiambaretto (Groupe Sup de Co Montpellier (GSCM) - Montpellier Business School) |
| Date: | 2021–03–01 |
| URL: | http://d.repec.org/n?u=RePEc:hal:journl:hal-03514785&r=dcm |
| By: | Stéphane Bonhomme; Kevin Dano; Bryan S. Graham |
| Abstract: | We study identification in a binary choice panel data model with a single predetermined binary covariate (i.e., a covariate sequentially exogenous conditional on lagged outcomes and covariates). The choice model is indexed by a scalar parameter θ, whereas the distribution of unit-specific heterogeneity, as well as the feedback process that maps lagged outcomes into future covariate realizations, are left unrestricted. We provide a simple condition under which θ is never point-identified, no matter the number of time periods available. This condition is satisfied in most models, including the logit one. We also characterize the identified set of θ and show how to compute it using linear programming techniques. While θ is not generally point-identified, its identified set is informative in the examples we analyze numerically, suggesting that meaningful learning about θ is possible even in short panels with feedback. |
| JEL: | C23 |
| Date: | 2023–03 |
| URL: | http://d.repec.org/n?u=RePEc:nbr:nberwo:31027&r=dcm |
| By: | Hammitt, James K.; Tuncel, Tuba |
| Abstract: | Individuals’ monetary values of decreases in mortality risk depend on the magnitude and timing of the risk reduction. We elicited stated preferences among three time paths of risk reduction yielding the same increase in life expectancy (decreasing risk for the next decade, subtracting a constant from or multiplying risk by a constant in all future years) and willingness to pay (WTP) for risk reductions differing in timing and life-expectancy gain. Respondents exhibited heterogeneous preferences over the alternative time paths, with almost 90 percent reporting transitive orderings. WTP is statistically significantly associated with life-expectancy gain (between about 7 and 28 days) and with respondents’ stated preferences over the alternative time paths. Estimated value per statistical life year (VSLY) can differ by time path and averages about $500, 000, roughly consistent with conventional estimates obtained by dividing estimated value per statistical life by discounted life expectancy. |
| Keywords: | value per statistical life: value per statistical life year; mortality risk, stated preference |
| JEL: | D61 I18 Q51 |
| Date: | 2023–03–10 |
| URL: | http://d.repec.org/n?u=RePEc:tse:wpaper:127946&r=dcm |
| By: | Josè L. Moraga González (Vrije Universiteit Amsterdam); Zsolt Sándor (Sapientia Hungarian University); Matthijs Wildenbeest (University of Arizona) |
| Abstract: | We propose a tractable method for estimation of a simultaneous search model for differentiated products that allows for observed and unobserved heterogeneity in both preferences and search costs. We show that for type I extreme value distributed search costs, expressions for search and purchase probabilities can be obtained in closed form. We show that our search model belongs to the generalized extreme value (GEV) class, which implies that it has a full information discrete-choice equivalent, and hence search data are necessary to distinguish between the search model and the equivalent full information model. We allow for price endogeneity when estimating the model and show how to obtain parameter estimates using a combination of aggregate market share data and individual level data on search and purchases. To deal with the dimensionality problem that typically arises in search models due to a large number of consideration sets we propose a novel Monte Carlo estimator for the search and purchase probabilities. Monte Carlo experiments highlight the importance of allowing for sufficient consumer heterogeneity when doing policy counterfactuals and show that our Monte Carlo estimator is accurate and computationally fast. Finally, a behavioral assumption on how consumers search provides a micro-foundation for consideration probabilities widely used in the literature. |
| Keywords: | demand estimation, price endogeneity, simultaneous search, differentiated products |
| JEL: | C14 D83 L13 |
| Date: | 2023–03–20 |
| URL: | http://d.repec.org/n?u=RePEc:tin:wpaper:20230015&r=dcm |
| By: | Koji Yokote; Isa E. Hafalir; Fuhito Kojima; M. Bumin Yenmez |
| Abstract: | Path independence is arguably one of the most important choice rule properties in economic theory. We show that a choice rule is path independent if and only if it is rationalizable by a utility function satisfying ordinal concavity, a concept closely related to concavity notions in discrete mathematics. We also provide a representation result for choice rules that satisfy path independence and the law of aggregate demand. |
| Date: | 2023–03 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2303.00892&r=dcm |
| By: | Faten Saliba; Giorgio Maarraoui; Walid Marrouch; Ada Wossink |
| Abstract: | This paper uses life satisfaction data to help the design of climate mitigation policies in the United Kingdom. We assess the effects of the exposure to ambient pollutants on long-term life satisfaction and short-term mental health in the UK. We estimate augmented Cobb-Douglas utility functions using pooled and random effects ordinal logit models. Results show that increases in NO2, PM10 and PM2.5 significantly decrease the odds of longterm happiness and short-term mental health in the UK. The willingness to pay for clean air is also significant and increases with level of education. These measurements derived can be used as benchmarks for pollution abatement subsidies or pollution taxes and can help in projecting a more comprehensive assessment of costs and benefits. |
| Keywords: | Air Pollution; Happiness; Policy Valuation; Climate Change; Environmental Policies; Pollution Taxes; Pollution Abatement Subsidies; life satisfaction data; dataset description; air pollutant; pollutants' correlation; household data; ordinal Logit; Income; Europe; Global |
| Date: | 2023–02–17 |
| URL: | http://d.repec.org/n?u=RePEc:imf:imfwpa:2023/035&r=dcm |
| By: | Stuart Donovan (Vrije Universiteit Amsterdam); Thomas de Graaff (Vrije Universiteit Amsterdam); Henri L.F. de Groot (Vrije Universiteit Amsterdam) |
| Abstract: | Using commuting data for Brisbane, Australia, we find that accounting for measurement error in travel times causes the magnitude of parameters in mode and location choice models to increase approximately three-fold and 30–40%, respectively. Errors appear to be somewhat systematic, with travel times being underestimated for short journeys and vice versa for long journeys—especially by public transport. We find similar results when we use alternative transport cost measures and independent commuting data from London. Our findings are likely to have important implications for transport and land use policy as well as the many types of economic models in which travel times—and transport costs, more generally—occupy a central role. |
| Keywords: | mode choice, location choice, travel times, measurement error, Australia |
| JEL: | C13 R14 R41 |
| Date: | 2023–03–06 |
| URL: | http://d.repec.org/n?u=RePEc:tin:wpaper:20230011&r=dcm |
| By: | Gabriele Camera (Chapman University); Jaehong Kim (Xiamen University); David Rojo Arjona (Chapman University) |
| Abstract: | Understanding how incentives and institutions help scaling up cooperation is important, especially when strategic uncertainty is considerable. Evidence suggests that this is challenging even when full cooperation is theoretically sustainable thanks to indefinite repetition. In a controlled social dilemma experiment, we show that adding partial cooperation choices to the usual binary choice environment can raise cooperation and efficiency. Under suitable incentives, partial cooperation choices enable individuals to cheaply signal their desire to cooperate, reducing strategic uncertainty. The insight is that richer choice sets can form the basis of a language meaningful for coordinating on cooperation. |
| Keywords: | experiments, repeated games, social dilemmas, strategy estimation |
| JEL: | C70 C90 D03 E02 |
| Date: | 2023 |
| URL: | http://d.repec.org/n?u=RePEc:chu:wpaper:23-05&r=dcm |
| By: | Joel L. Horowitz; Ahnaf Rafi |
| Abstract: | We consider penalized extremum estimation of a high-dimensional, possibly nonlinear model that is sparse in the sense that most of its parameters are zero but some are not. We use the SCAD penalty function, which provides model selection consistent and oracle efficient estimates under suitable conditions. However, asymptotic approximations based on the oracle model can be inaccurate with the sample sizes found in many applications. This paper gives conditions under which the bootstrap, based on estimates obtained through SCAD penalization with thresholding, provides asymptotic refinements of size \(O \left( n^{- 2} \right)\) for the error in the rejection (coverage) probability of a symmetric hypothesis test (confidence interval) and \(O \left( n^{- 1} \right)\) for the error in rejection (coverage) probability of a one-sided or equal tailed test (confidence interval). The results of Monte Carlo experiments show that the bootstrap can provide large reductions in errors in coverage probabilities. The bootstrap is consistent, though it does not necessarily provide asymptotic refinements, even if some parameters are close but not equal to zero. Random-coefficients logit and probit models and nonlinear moment models are examples of models to which the procedure applies. |
| Date: | 2023–03 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2303.09680&r=dcm |
| By: | Joel L. Horowitz; Ahnaf Rafi |
| Abstract: | We consider penalized extremum estimation of a high-dimensional, possibly nonlinear model that is sparse in the sense that most of its parameters are zero but some are not. We use the SCAD penalty function, which provides model selection consistent and oracle efficient estimates under suitable conditions. However, asymptotic approximations based on the oracle model can be inaccurate with the sample sizes found in many applications. This paper gives conditions under which the bootstrap, based on estimates obtained through SCAD penalization with thresholding, provides asymptotic refinements of size O (n−2) for the error in the rejection (coverage) probability of a symmetric hypothesis test (confidence interval) and O (n−1) for the error in rejection (coverage) probability of a one-sided or equal tailed test (confidence interval). The results of Monte Carlo experiments show that the bootstrap can provide large reductions in errors in coverage probabilities. The bootstrap is consistent, though it does not necessarily provide asymptotic refinements, even if some parameters are close but not equal to zero. Random-coefficients logit and probit models and nonlinear moment models are examples of models to which the procedure applies. |
| Date: | 2023–03–29 |
| URL: | http://d.repec.org/n?u=RePEc:azt:cemmap:06/23&r=dcm |
| By: | Gustavo F. Serenelli; Emiliano Delfau |
| Abstract: | The goal of this document is to present a methodology for estimating probabilities for ordered sets. This may have several practical applications such as calibration of Rating Models, estimation of Mortality Tables or measurement of side effects related to different doze sizes. In order to do this, an Objective / Non Informative Bayesian approach is applied, through which, using a multidimensional Jeffreys prior, a posterior distribution may be inferred for each of the probabilities being estimated |
| Keywords: | Bayesian estimation, probability estimation, uninformative prior, rating calibration, low default, ordered sets, jeffreys prior. |
| Date: | 2023–03 |
| URL: | http://d.repec.org/n?u=RePEc:cem:doctra:849&r=dcm |
| By: | Gustafson, Christopher R. (University of Nebraska-Lincoln); Meerza, Syed Imran Ali |
| Abstract: | Experimental auctions are an important technique for measuring preferences for products, product attributes, and the impact of information. While these techniques are widely used, there is a paucity of evidence about an important design decision available to guide researchers: the choice of a between-subject vs. within-subject design. Within-subject designs offer clear value in terms of providing multiple observations per participant, which increases statistical power, but there are long-standing concerns about properties that could decrease the external validity of results generated in within-subject experiments. In this paper, we examine the impact of information on the economically motivated mislabeling of extra virgin olive oil (EVOO) on consumer valuation of EVOOs produced in the country that has experienced mislabeling scandals, along with EVOOs from two unimplicated countries, in between-subject and within-subject designs. Our findings show that the significance and relative impacts on valuation are identical between the two conditions. In fact, the valuation of the implicated EVOO differs only by a few cents ($3.53 vs. $3.60) after participants received information about mislabeling. There were larger differences in valuation of the two unimplicated EVOOs post-information, though the relative preferences implied by the results and the statistical significance did not differ between the conditions. The impacts of information on the product “targeted” by the information are measured consistently in both between-subjects and within-subjects designs, while we observe more variation in off-target products, suggesting that the researchers who are interested in informational spillovers may need to be more careful in design choice than those who want to estimate the impact of information on the target products. |
| Date: | 2023–03–16 |
| URL: | http://d.repec.org/n?u=RePEc:osf:osfxxx:3g4m5&r=dcm |
| By: | Astrid Dannenberg (University of Kassel); Eva Weingaertner (University of Kassel) |
| Abstract: | Our choice of food has major impacts on the environment. At the same time, it is visible to all people with whom we spend our daily lives. This raises the question of whether people are adapting their diets to gain a green reputation, as has been observed for other environmentally relevant consumption choices. Using an experiment in which participants can choose between vegan, vegetarian, and meat-based food vouchers, we examine how observation by others and the provision of an information nudge influence food choices. The results show that providing an information nudge reduces the likelihood of choosing meat by 12 percentage points. Observation by others does not significantly reduce the likelihood of choosing meat. Contrary to our prediction, when participants are observed and receive the information nudge, they are less inclined to choose one of the more sustainable options. We discuss the reasons for the partly surprising results and the implications for policy. |
| Keywords: | Food choice; meat consumption; information nudge; observability; experiment |
| JEL: | C9 D91 Q18 |
| Date: | 2023 |
| URL: | http://d.repec.org/n?u=RePEc:mar:magkse:202301&r=dcm |
| By: | Agustín Cabrera; Alejandro D. Pereyra; Gustavo Luis Demarco |
| Keywords: | Deuda Pública, Probabilidad de impago, Flujos de capitales |
| JEL: | H63 |
| Date: | 2021–11 |
| URL: | http://d.repec.org/n?u=RePEc:aep:anales:4446&r=dcm |