|
on Discrete Choice Models |
| By: | Xuan Di; Henry X. Liu; Shanjiang Zhu; David Levinson (Nexus (Networks, Economics, and Urban Systems) Research Group, Department of Civil Engineering, University of Minnesota) |
| Abstract: | The replacement I-35W bridge in Minneapolis saw less traffic than the original bridge though it provided substantial travel time saving for many travelers. This observation cannot be explained by the classical route choice assumption that travelers always take the shortest path. Accordingly, a boundedly rational route switching model is proposed assuming that travelers will not switch to the new bridge unless travel time saving goes beyond a threshold or “indifference bandâ€. To validate the boundedly rational route switching assumption, route choices of 78 subjects from a GPS travel behavior study were analyzed before and after the addition of the new I-35W bridge. Indifference bands are estimated for both commuters who were previously bridge users and those who never had the experience of using the old bridge. This study offers the first empirical estimation of bounded rationality parameters from GPS data and provides guidelines for traffic assignment. |
| Keywords: | Route Choice, Travel Demand Modeling, Bounded Rationality, Indifference Band, GPS Study, Travel Behavior, Networks |
| JEL: | R40 |
| Date: | 2014 |
| URL: | http://d.repec.org/n?u=RePEc:nex:wpaper:indifferencebandsforrouteswitching&r=dcm |
| By: | Hamna Ahmed (Lahore School of Economics, Lahore, Pakistan.); Sahar Amjad (Lahore School of Economics, Lahore, Pakistan.); Masooma Habib (Lahore School of Economics, Lahore, Pakistan.); Syed Ahsan Shah (Lahore School of Economics, Lahore, Pakistan.) |
| Abstract: | The objective of this study is to understand why parents in rural areas choose low-cost private schools when free public schools are available. The study employs data from the Privatization in Education Research Initiative (PERI) School Choice Survey, 2011. The sample under study comprises 5–18-year-old children enrolled in private or public schools at the primary, secondary, or high school level in eight rural tehsils across Punjab, Pakistan. Our methodology entails specifying a probability choice model to understand what determines school choice in a rural setting. The variable of interest is parents’ perceptions of their child’s competence, the quality of the child’s school, and the employment opportunities available to the child. The model also controls for a range of child-, parent-, and household-specific characteristics. Five main factors emerge as important determinants of private school choice. These include the socioeconomic status of the household, the degree of a school’s accessibility, the cost of schooling, parents’ perceptions of school quality, and their perceptions of the available employment opportunities in the region. |
| Keywords: | School choice, private, public, perceptions, school quality, employment opportunities, wealth, access, cost of schooling, Punjab, Pakistan. |
| JEL: | I21 I25 O12 |
| Date: | 2013 |
| URL: | http://d.repec.org/n?u=RePEc:lje:wpaper:1-2013&r=dcm |
| By: | Francisco Javier Eransus (Departamento de Economía Cuantitativa (Department of Quantitative Economics), Facultad de Ciencias Económicas y Empresariales (Faculty of Economics and Business), Universidad Complutense de Madrid); Alfonso Novales Cinca (Departamento de Economía Cuantitativa (Department of Quantitative Economics), Facultad de Ciencias Económicas y Empresariales (Faculty of Economics and Business), Universidad Complutense de Madrid) |
| Abstract: | We analyze the effect of parameter estimation error on the size of unconditional population level tests of predictive ability when they are implemented under a class of loss functions we refer to as ‘discrete functions’. The analysis is restricted to linear models in stationary variables. We obtain analytical results for no nested models guaranteeing asymptotic irrelevance of parameter estimation error under a plausible predictive environment and three subsets of discrete loss functions that seem quite appropriate for many economic applications. For nested models, we provide some Monte Carlo evidence suggesting that the asymptotic distribution of the Diebold and Mariano (1995) test is relatively robust to parameter estimation error in many cases if it is implemented under discrete loss functions, unlike what happens under the squared forecast error or the absolute value error loss functions. |
| Keywords: | Parameter uncertainty; Forecast accuracy; Discrete loss function. |
| JEL: | C53 C52 C12 |
| Date: | 2014 |
| URL: | http://d.repec.org/n?u=RePEc:ucm:doicae:1422&r=dcm |
| By: | Moussa Dieng (CERDI - Centre d'études et de recherches sur le developpement international - CNRS : UMR6587 - Université d'Auvergne - Clermont-Ferrand I); Martine Audibert (CERDI - Centre d'études et de recherches sur le developpement international - CNRS : UMR6587 - Université d'Auvergne - Clermont-Ferrand I); Jean-Yves Le Hesran (IRD - IRD - Université Descartes); Anta TA DIAL (ISED - ISED - Université Cheik Anta Diop) |
| Abstract: | Depuis les années 2000, le Sénégal a adopté des politiques nationales visant la suppression progressive du paiement direct au point de services pour rendre les soins de santé plus accessibles. La mise en place de ces politiques de subvention et de gratuité dans un espace dense hétérogène voire hétéroclite, présente une situation particulière. Pour comprendre ces interactions et étudier le comportement des ménages en matière de demande de soins, 5520 individus ont étés enquêtés à quatre reprises sur la période 2010-2011 dans la banlieue de Dakar (Pikine), un probit multinomial est estimé pour étudier la demande de soins de la population face à un épisode de maladie. Les résultats montrent que l'effet négatif du prix est en moyenne assez faible, mais qu'il varie en fonction du niveau de revenu et de la sévérité de la maladie. La qualité perçue des soins a un effet positif sur le recours aux services de santé privés pour lesquels on observe une compensation de l'effet négatif du prix par la qualité. L'effet de l'âge n'est pas linéaire et les enfants, plus touchés par la maladie, bénéficient de peu d'exemption ou du moins d'exemption partielle contrairement aux personnes âgées qui bénéficient d'exemption totale (plan SESAME).// Summary // Since 2000, Senegal has adopted national policies aimed at phasing out direct payments made at the point of service to make health care more accessible to the populations. The implementation of these policies subsidizing health care is particularly challenging in an environment with high density and important mix of population such as the suburbs of Dakar. To understand the underlying interactions and study the behavior of households' demand for care, 5520 individuals were surveyed four times over the period December 2010 to November 2011 in Pikine, one of the main suburbs of Dakar. A multinomial probit was estimated to assess the determinants of health care demand in an episode of illness. The results showed a negative effect of price but with a low coefficient. The magnitude of the price effect varies depending on individual income level and the severity of the disease. The quality of care (self-assessed by individuals) has a positive effect on the use of private health services and seems to compensate the negative effect of price in that case. The effect of age is not linear and children, who are most affected by diseases, receive no exemption or at most partial exemption compared to seniors who receive full exemption (SESAME plan). |
| Keywords: | demande de soins;subvention des soins;probit multinomial;Pikine;senegal |
| Date: | 2014–07–22 |
| URL: | http://d.repec.org/n?u=RePEc:hal:wpaper:halshs-01027504&r=dcm |
| By: | Tocco, Barbara; Bailey, Alastair; Davidova, Sophia |
| Abstract: | This paper presents an empirical methodology for studying the reallocation of agricultural labour across sectors from micro data. Whereas different approaches have been employed in the literature to better understand the mobility of labour, looking at the determinants to exit farm employment and enter off-farm activities, the initial decision of individuals to work in agriculture, as opposed to other sectors, has often been neglected. The proposed methodology controls for the selectivity bias, which may arise in the presence of a non-random sample of the population, in this context those in agricultural employment, which would lead to biased and inconsistent estimates. A 3-step multivariate probit with two selection and one outcome equations constitutes the selected empirical approach to explore the determinants of farm labour to exit agriculture and switch occupational sector. The model can be used to take into account the different market and production structures across European member states on the allocation of agricultural labour and its adjustments. |
| Keywords: | Farm Labour, Reallocation, Micro-Data, Labor and Human Capital, |
| Date: | 2013–05 |
| URL: | http://d.repec.org/n?u=RePEc:ags:famawp:155703&r=dcm |
| By: | Pilar Abad (Universidad Rey Juan Carlos, Paseo Artilleros s/n, 28032 Madrid, Spain and RFA-IREA.); M. Dolores Robles (Universidad Complutense de Madrid and ICAE, Campus de Somosaguas, 28223 Madrid, Spain.) |
| Abstract: | Risk-averse investors take into consideration risk-return tradeoff for decide their new position after the release of relevant information. This paper analyzes the informational content of rating change announcements focusing on the joint reaction they cause on the risk-return binomial. Our purpose is to identify the main factors that signal which announcements are informative. To do that we estimate a binomial logit model for the probability of informative content of credit rating announcements. We analyze a sample of rating events affecting Spanish listed firms from 2000 to 2010. Empirical results show significant differences in the informative content between positive and negative rating events. For both kinds of announcements, we find higher informative content when agencies agree about the new level of solvency, whereas those affecting high covered firm that operate in highly regulated sectors are the less informative. Other factors as the presence of a previous rating refinements or trends in the credit quality reveals different information depending on the direction of the rating event. Finally, we find the announcements after de crisis disclose less information, suggesting a loss of reputation of CRAs. |
| Keywords: | Abnormal return, Abnormal systematic risk, Abnormal volatility, Logit model. |
| JEL: | G12 G14 G24 C22 |
| Date: | 2014–07 |
| URL: | http://d.repec.org/n?u=RePEc:ucm:doicae:1423&r=dcm |