|
on Discrete Choice Models |
| By: | Janssen, Ingrid; Dijkman, Wouter; Heij, Tim Op; Willems, Rick; Borgers, Aloys |
| Abstract: | Dutch inner-city shopping areas face a decreasing number of visitors and declining sales volumes. Internet shopping, economic decline and an ageing population are considered to be the main causes. Improving the atmospherics of inner-city shopping areas may be a solution to improve the competitiveness relative to other recreational destinations and attract more visitors to shopping centres. The aim of this research is to empirically determine which atmospheric characteristics contribute to the shopper's appreciation of inner-city shopping areas. A list of 25 environmental characteristics formed the basis of a survey that was conducted in the historic inner-city shopping areas of two Dutch medium sized cities: Maastricht and 's-Hertogenbosch. Within each of these inner-city areas, four locations were selected; two historical and two non-historical. At each location, the 25 characteristics were firstly assessed by the researchers. Secondly, in total 918 consumers were invited to rate each characteristic. In addition, each respondent rated the overall appreciation and the sphere of the location under consideration. Furthermore, each respondent was asked to rank order the four locations of the particular inner-city considering the most favorite shop location and the most atmospheric shop location. Decision tree analysis was used to find out if and which physical characteristics of the locations cause the largest impact on the consumers' appreciations. Furthermore, multinomial logit (MNL) choice models were estimated to predict the most favorite and most atmospheric locations from the rank orderings provided by each respondent. This analysis showed which combination of observed characteristics contributed most to the respondent's first choice of favorite location and the respondent's first choice of most atmospheric location. Although the explanatory power of the models is limited, some observed characteristics appeared to significantly influence the shoppers' preferences. The results showed that the most favorite shopping area does not need to be the most atmospheric shopping area. Nonetheless, adjusting shopping areas to the atmospheric characteristics that contribute significantly to consumer's appreciation will help shopping center managers and developers to improve the competing power of their shopping areas. |
| Date: | 2013 |
| URL: | http://d.repec.org/n?u=RePEc:arz:wpaper:eres2013_96&r=dcm |
| By: | Gomes, Sandra Vieira |
| Abstract: | The objective of this paper is to provide an empirical analysis of the influence of traffic noise has on the housing market. Traffic noise can be considered as an environmental externality from road transport system. As so, it affects the value that people are willing to pay for their houses. The value of this effect has already been studied by some authors, with Stated Preference-choice (SP-choice) methods, the hedonic pricing model, or others. The approach used in this study combines traditional methods, namely linear modelling techniques used for quantifying the effect on housing prices fluctuation, with newer methods as the spatial analysis associate d to Geographical Information Systems. There are obvious advantages to a GIS in adding spatial analytical capabilities in terms of increasing its functionality and meeting a demand for systems that do something beyond storing, retrieving and displaying large amounts of information. This study was developed for the city of Lisbon, Portugal. A geographical database was created combining information from the Census, on the housing characteristics, with renting and sales activities. This study applies spatial auto-correlation techniques to analyze the relation between traffic noise and the variation on the housing value in Lisbon, and aims to contribute to the scientific knowledge of housing market in Lisbon, with detailed and accurate information on the factors affecting its variation. |
| Date: | 2014 |
| URL: | http://d.repec.org/n?u=RePEc:arz:wpaper:eres2014_92&r=dcm |
| By: | Yeh, Kuang-Yih; Hsia, Hao-Ching |
| Abstract: | The dramatic growing of private vehicles ownership results in heavy traffic congestion and serious environmental pollution in the city, such as air pollution and annoying noise. Professional urban planning experts have attempted to achieve the target mentioned above by adjusting current land use. However, adjusting land use always takes a long time. At the moment, the advanced countries in Europe and the United States have been already paying attention to traffic demand management (TDM) in order to solve these environmental problems and achieve the target of building a sustainable living environment. While road pricing (RP) is generally regarded as one of the most effective measures of TDM, its poor acceptability has been the greatest impediment to its implementation. Japanese scholars proposed the parking deposit system (PDS) as an alternative RP scheme to improve the public acceptance. In view of the above, this study takes a local commercial district of Tainan City as the study area where the data of visitors driving private vehicle are collected by a stated preference questionnaire. Because the acceptances of two policies are considered to be correlated in this study, the choice behavior model of acceptance is established by using bivariate binary probit model. This study also considered that different groups may hold different attitudes, so respondents were separated into two groups according to their consciousness of fairness. Therefore, the choice behavior model of acceptance of fair group and unfair group are built and compared. The result of model estimation has several good implications for the proposal of environmental policy and traffic demand management measures. It indicates that the most important thing that people really care about is cost. That means individual’s behavior can be adjusted by charge schemes. On the other hand, the result of social interaction equilibrium shows that the difference can be obviously distinguished from different groups, so it is recognized that the approval probability of the fair group is higher than the unfair group. The marginal effects of estimated coefficients are also discussed in this study. |
| Date: | 2014 |
| URL: | http://d.repec.org/n?u=RePEc:arz:wpaper:eres2014_192&r=dcm |
| By: | Matthew J. Baker (Hunter College and Graduate Center, City University of New York) |
| Abstract: | I describe algorithms for drawing from distributions using adaptive Markov chain Monte Carlo (MCMC) methods, introduce a Mata function for performing adaptive MCMC, amcmc(), and a suite of functions amcmc *() allowing an alternative implementation of adaptive MCMC. amcmc() and amcmc *() may be used in conjunction with models set up to work with Mata’s [M-5] moptimize( ) or [M-5] optimize( ), or with stand-alone functions. To show how the routines might be used in estimation problems, I give two examples of what Chernozukov and Hong (2003) refer to as Quasi-Bayesian or Laplace-Type estimators - simulation-based estimators employing MCMC sampling. In the first example I illustrate basic ideas and show how a simple linear model can be estimated by simulation. In the next example, I describe simulation-based estimation of a censored quantile regression model following Powell (1986); the discussion describes the workings of the Stata command mcmccqreg. I also present an example of how the routines can be used to draw from distributions without a normalizing constant, and in Bayesian estimation of a mixed logit model. This discussion introduces the Stata command bayesmlogit. |
| Keywords: | Stata, Mata, Markov chain Monte Carlo, drawing from distributions, mixed logit Bayesian estimation, bayesmlogit, mcmccqreg |
| JEL: | C10 C11 C13 C15 C25 C60 |
| Date: | 2014–07–18 |
| URL: | http://d.repec.org/n?u=RePEc:cgc:wpaper:003&r=dcm |