|
on Discrete Choice Models |
| By: | Adhya Sumanta; Banerjee Tathagata; Chattopadhyay Gaurangadeb |
| Abstract: | We consider the estimation of finite population proportions of categorical survey responses obtained by probability sampling. The customary design-based estimator does not make use of the auxiliary data available for all the population units at the estimation stage. We adopt a model-based predictive approach to incorporate this information and make the estimates more efficient. In the first part of our paper we consider a multinomial logit type model when logit function is a known parametric function of the covariates. We then use it for the prediction of non-sampled responses. This together with sampled responses is used to obtain the estimates of the proportions. The asymptotic biases and variances of these estimators are obtained. The main drawback of this approach is, being a parametric model it may suffer from model misspecification and thus, may lose it’s efficiencies over the usual design-based estimates. To overcome this drawback, in the next part of this paper we replace the multinomial logit type model by a nonparametric model using recently developed random coefficients splines models. Finally, we carry out a simulation study. It shows that the nonparametric approach may lead to an appreciable improvement over both parametric and design-based approaches when the regression function is quite different from multinomial logit. |
| Keywords: | Auxiliary information, Model-based inference, Finite population estimation, Multinomial logit, Random coefficients splines models, Laplace approximation |
| Date: | 2007–05–14 |
| URL: | http://d.repec.org/n?u=RePEc:iim:iimawp:2007-05-07&r=dcm |
| By: | Angela Robinson (University of East Anglia); Judith Covey (University of Durham); Anne Spencer (Queen Mary, University of London); Graham Loomes (University of East Anglia) |
| Abstract: | Previous research has shown that people wish a premium to be placed on the prevention of certain types of deaths as they perceive those deaths as ‘worse’ than others. The research reported in this paper is an attempt to quantify such a ‘bad death’ premium via a discrete choice experiment (DCE). The four underlying attributes included were: the age of the victim, who was most to blame for the death, the severity of the victim’s pain and suffering in the period leading up to death, and the duration of the victim’s pain and suffering in the period leading up to death. In addition, a fifth attribute - number of deaths- was included in order to provide a <i>quantitative</i> scale against which to measure the “bad death premium”. The results show that each of the 4 underlying attributes did matter to respondents in determining whether deaths were worse than others, but also uncovered marked insensitivity to variations in the number of those deaths. The implication of our findings for the use of quantitative variables in DCEs is discussed. |
| Keywords: | Discrete choice experiment, Value of preventing a fatality, Relative weights, Insensitivity |
| JEL: | H5 I10 |
| Date: | 2007–05 |
| URL: | http://d.repec.org/n?u=RePEc:qmw:qmwecw:wp597&r=dcm |
| By: | Arimura, Toshi (Resources for the Future); Hibiki, Akira; Katayama, Hajime |
| Abstract: | Using Japanese facility-level data from an Organisation for Economic Co-operation and Development survey, we estimate the effects of implementation of ISO14001 and publication of environmental reports on the facilities’ environmental performance. While most previous studies focused on an index of emissions toxicity, this study examines three areas of impacts, none of which have been explored in the literature: natural resource use, solid waste generation, and wastewater effluent. The study is also unique in that the effectiveness of ISO14001 is considered in relation to environmental regulations. Our findings are summarized as follows. First, both ISO14001 and report publication help reduce all three impacts; the former appears more effective in all areas except wastewater. Second, environmental regulations do not weaken the effect of ISO14001. Third, assistance programs offered by local governments—a voluntary approach—promote facilities’ adoption of ISO14001. These findings suggest that governments can use command-and-control and voluntary approaches concurrently. |
| Keywords: | environmental management systems, ISO14001, environmental reports, voluntary actions, governmental assistance programs, environmental impacts, discrete choice model, endogeneity, GHK simulator |
| JEL: | C35 Q53 Q58 |
| Date: | 2007–05–16 |
| URL: | http://d.repec.org/n?u=RePEc:rff:dpaper:dp-07-31&r=dcm |
| By: | Colignatus, Thomas |
| Abstract: | Democratic nations are advised to have parliaments select the chief executive by the Borda Fixed Point method. The current practice of having direct popular elections using systems that have originated in history is inoptimal and actually quite undemocratic since winners are selected who don’t reflect the national sentiment. The paper gives the example of the 2007 French presidential elections. Under the currently and historically grown system of run-off plurality Sarkozy got elected while the more democratic method of Borda Fixed Point would have generated Bayrou. The example uses reasonable assumptions on underlying micro preferences. |
| Keywords: | voting theory; voting systems; elections; public choice; political economy; run-off plurality; Borda Fixed Point; democracy |
| JEL: | D71 H0 A2 |
| Date: | 2007–05–10 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:3170&r=dcm |
| By: | Junyi Shen (OSIPP,Osaka University); Tatsuyoshi Saijo (ISER,Osaka University) |
| Abstract: | In this paper we apply hypothetical choice experiments through a field survey in Shanghai of China to examine whether China Energy Efficiency Label affects consumersf choices of air conditioner and refrigerator. A latent class approach is used to observe both heterogeneities among the respondents and product brands. The results suggest that the effect of energy efficiency label on consumersf preferences is twofold. First, more energy efficient air conditioners or refrigerators are preferred by consumers, no matter whether they are with foreign brands or domestic brands and whether they are new or second-hand. Second, energy efficiency label per se is recognized by consumers. In addition, presence of a (hypothetical) label that indicates the electricity billfs difference comparing to a standard model is significantly preferred by the respondents in most of the cases, suggesting that more information provided to consumers makes them much happier. Finally, the class probability weighted willingness to pay values for one rank upgrading in energy efficiency of refrigerator are higher than those of air conditioner, implying that consumers have an incentive to pay more for appliances used more frequently. |
| Keywords: | Energy efficiency label, Consumersf purchase decision, Latent class model, Willingness to pay, China |
| JEL: | C25 C93 D12 Q49 |
| Date: | 2007–05 |
| URL: | http://d.repec.org/n?u=RePEc:osp:wpaper:07e005&r=dcm |