|
on Discrete Choice Models |
| By: | Valerio Gatta (University of Roma Tre and CREI); Edoardo Marcucci (University of Roma Tre and CREI) |
| Abstract: | This paper jointly investigates non-linear attribute effects and discrete mixture heterogeneity. The research context relates to urban freight transport policy evaluation. The paper adopts an agent-specific perspective. The often unforeseen and undesired results deriving from urban freight transport policy implementation have induced many researchers to call for an in-depth analysis of specific agents' preferences. However, the structural lack of appropriate data has hindered investigations at such a detailed level. This paper contributes to bridging this data and knowledge gap by: constructing an original data set; testing for non-linear effects in attribute level variations; investigating the presence of inter and intra-agent heterogeneity; jointly exploring non-linear attribute effects and intraagent heterogeneity. The results obtained underline the relevance of intra-agent preference heterogeneity and non-linear effects of attribute variations. More in detail, the paper detects two classes of agents with substantially different preferences with respect to the possible policy interventions. Non-linear sensitivity suggests policy makers should carefully consider the effects induced by the specific status quo level for policy relevant attribute variations. Intraagent discrete heterogeneity implies different willingness to pay measures for given policy changes. The presence of both non-linear effects and intra-agent heterogeneity suggests policy makers to explicitly consider the status quo level that is to be changed while, at the same time, contemplate differentiated reactions deriving from the implementation of a urban freight policy change. In conclusion, the paper underlines the need for rigorous ex-ante policy analysis if the correct policy outcomes are to be estimated with an adequate level of accuracy. |
| Keywords: | Urban freight transport, discrete mixture heterogeneity, non-linear attribute effects, policy evaluation |
| Date: | 2013 |
| URL: | http://d.repec.org/n?u=RePEc:rcr:wpaper:03_13&r=dcm |
| By: | Paolo Delle Site (DICEA Dept. of Civil Architectural and Environmental Engineering, CTL Centre of Research on Transport and Logistics, University of Rome La Sapienza) |
| Abstract: | We provide the discrete choice, random utility counterparts of some basic results of consumer theory. For the primal problem and related Marshallian probabilities, we provide a new, simpler proof of Roy's identity at aggregate level and investigate price and income effects. For the dual problem and related Hicksian probabilities, we extend Shepard's lemma at aggregate level to unbound expenditure and investigate compensated price effects. We establish a primal-dual equivalence result and provide the counterpart of the Slutsky equation. |
| Date: | 2013 |
| URL: | http://d.repec.org/n?u=RePEc:rcr:wpaper:05_13&r=dcm |
| By: | Li, Xiaogu; Clark, Christopher D; Jensen, Kimberly L; Yen, Steven T |
| Abstract: | This study examines how a $50 mail-in rebate influences consumer willingness-to-pay for an ENERGY STAR-certified refrigerator. Data collected from a 2009 U.S. online survey containing a hypothetical choice experiment. Results suggest that a rebate induces uncertainty about the quality of ENERGY STAR-certified refrigerators and, thus, could actually reduce willingness-to-pay. |
| Keywords: | Choice Experiment, Eco-label, Energy Star, Generalized Multinomial Logit, Ordered Probit, Rebate, Refrigerator, Willingness-to-Pay, Environmental Economics and Policy, Q58, D12, |
| Date: | 2013 |
| URL: | http://d.repec.org/n?u=RePEc:ags:saea14:159795&r=dcm |
| By: | Jakob W. Messner; Georg J. Mayr; Daniel S. Wilks; Achim Zeileis |
| Abstract: | Extended logistic regression is a recent ensemble calibration method that extends logistic regression to provide full continuous probability distribution forecasts. It assumes conditional logistic distributions for the (transformed) predictand and fits these using selected predictand category probabilities. In this study we compare extended logistic regression to the closely related ordered and censored logistic regression models. Ordered logistic regression avoids the logistic distribution assumption but does not yield full probability distribution forecasts, whereas censored regression directly fits the full conditional predictive distributions. To compare the performance of these and other ensemble post-processing methods we used wind speed and precipitation data from two European locations and ensemble forecasts from the European Centre for Medium-Range Weather Forecasts (ECMWF). Ordered logistic regression performed similarly to extended logistic regression for probability forecasts of discrete categories whereas full predictive distributions were better predicted by censored regression. |
| Keywords: | probabilistic forecasting, extended logistic regression, ordered logistic regression, heteroscedasticity |
| JEL: | C53 C25 Q42 |
| Date: | 2013–10 |
| URL: | http://d.repec.org/n?u=RePEc:inn:wpaper:2013-32&r=dcm |
| By: | Nadja Klein; Thomas Kneib; Stephan Klasen; Stefan Lang |
| Abstract: | In this paper, we propose a unified Bayesian approach for multivariate structured additive distributional regression analysis where inference is applicable to a huge class of multivariate response distributions, comprising continuous, discrete and latent models, and where each parameter of these potentially complex distributions is modelled by a structured additive predictor. The latter is an additive composition of different types of covariate effects e.g. nonlinear effects of continuous variables, random effects, spatial variations, or interaction effects. Inference is realised by a generic, efficient Markov chain Monte Carlo algorithm based on iteratively weighted least squares approximations and with multivariate Gaussian priors to enforce specific properties of functional effects. Examples will be given by illustrations on analysing the joint model of risk factors for chronic and acute childhood malnutrition in India and on ecological regression for German election results. |
| Keywords: | correlated responses; iteratively weighted least squares proposal; Markov chain Monte Carlo simulation; penalised splines; semiparametric regression; Dirichlet regression; seemingly unrelated regression |
| Date: | 2013–11 |
| URL: | http://d.repec.org/n?u=RePEc:inn:wpaper:2013-35&r=dcm |
| By: | KUPFER, Franziska; KESSELS, Roselinde; GOOS, Peter; VAN DE VOORDE, Eddy; VERHETSEL, Ann |
| Abstract: | Airport competition is a topic which recently gained interest in transport research. However, many studies about airport competition focus on passengers or passenger operations. Research about airport competition for air cargo is still scarce. This paper contributes to the understanding of this topic by analyzing the airport choice for freighter operations in Europe. It first reveals the choice process that airports follow, as well as the different factors that play a role therein. Furthermore, using a discrete choice experiment, we analyzed six choice factors more in-depth. We collected completed questionnaires from 26 airlines and used the discrete choice data as input for a multinomial logit model. The results show that the presence of passenger operations at an airport is not a significant factor in explaining airlines’ choices, which, from an airline’s point of view, supports the idea of all-cargo airports and therefore the relocation of cargo operations to non-congested airports. The presence of forwarders, on the other hand, is the most important factor. This shows that, when trying to influence airlines in their airport choice, airports and policy makers also have to consider the preferences of forwarders. |
| Keywords: | Air cargo, Discrete choice analysis, Airport choice, Multinomial logit |
| Date: | 2013–11 |
| URL: | http://d.repec.org/n?u=RePEc:ant:wpaper:2013028&r=dcm |
| By: | Czajkowski, Mikolaj; Hanley, Nicholas; LaRiviere, Jacob |
| Abstract: | This paper develop and estimates a model of demand estimation for environmental public goods which allows for consumers to learn about their preferences through consumption experiences. We develop a theoretical model of Bayesian updating, perform comparative statics over the model, and show how the theoretical model can be consistently incorporated into a reduced form econometric model. We then estimate the model using data collected for two environmental goods. We find that the predictions of the theoretical exercise that additional experience makes consumers more certain over their preferences in both mean and variance are supported in each cas e. |
| Keywords: | scale variance; scale; generalized multinomial logit; Bayesian updating; stated preferences; preference learning; discrete choice experiment |
| Date: | 2013–10 |
| URL: | http://d.repec.org/n?u=RePEc:stl:stledp:2013-11&r=dcm |
| By: | Lei, J. (Tilburg University, Center for Economic Research) |
| Abstract: | Abstract: This paper considers spatial autoregressive (SAR) binary choice models in the context of panel data with fixed effects, where the latent dependent variables are spatially correlated. Without imposing any parametric structure of the error terms, this paper proposes a smoothed spatial maximum score (SSMS) estimator which consistently estimates the model parameters up to scale. The identification of parameters is obtained, when the disturbances are time-stationary and the explanatory variables vary enough over time along with an exogenous and time-invariant spatial weight matrix. Consistency and asymptotic distribution of the proposed estimator are also derived in the paper. Finally, a Monte Carlo study indicates that the SSMS estimator performs quite well in finite samples. |
| Keywords: | Spatial Autoregressive Models;Binary Choice;Fixed Effects;Maximum Score Estimation |
| JEL: | C14 C21 C23 C25 R15 |
| Date: | 2013 |
| URL: | http://d.repec.org/n?u=RePEc:dgr:kubcen:2013061&r=dcm |
| By: | Simone BERTOLI; Jes�s FERN�NDEZ-HUERTAS MORAGA; Herbert BR�CKER |
| Abstract: | The European crisis has diverted migration flows away from countries affected by the recession towards Germany. The diversion process creates a challenge for traditional discrete-choice models that assume that only bilateral factors account for dyadic migration rates. This paper shows how taking into account the sequential nature of migration decisions leads to write the bilateral migration rate as a function of expectations about the evolution of economic conditions in alternative destinations. Empirically, we incorporate 10-year bond yields as an explanatory variable capturing forward-looking expectations and apply our model to an empirical analysis of migration from the countries of the European Economic Association to Germany in the period 2006-2012. We show that disregarding alternative destinations leads to substantial biases in the estimation of the determinants of migration rates. |
| JEL: | J61 O15 F22 |
| Date: | 2013 |
| URL: | http://d.repec.org/n?u=RePEc:cdi:wpaper:1467&r=dcm |