Study On The Models Of Valuing Coastal Tourism Resources

Resource and environmental economics classifies resources/goods into two categories: one is market resource/good and the other is non-market resource/good. The value of market resource/good could be reflected via the price determined by market supply and demand, whereas, due to incomplete market environment for the non-market good and resource, the prices of those types of good and resource cannot be used to reflect their true economic value. In fact, under many circumstances, the prices of non-market resources may not exist. Nevertheless, the non-existence of price does not mean zero economic value to be accrued to those non-market resources, neither are their values very low. Observation that people exhibit a high demand for various beautiful natural environments, such as clean water and air, extraordinary sight of tourism sites suggests tremendous economic value that should be accrued to those non-market resource/good. One instrument which is commonly used to measure economic value for non-market resource is the consumer surplus hypothetically acted as a price. Thus, in this thesis the consumer surplus is utilized to measure economic value of non-market resources, including their use value and nonuse value as well. Taking coastal tourism in Dalian as an empirical example to assess the economic value for a variety of coastal environmental tourism resources and activities involves two major steps:the first step is to establish econometric models and estimate model parameters and following up by doing related statistical tests for the parameters. The second step is to use the estimated econometric model and coefficients to calculate economic value of coastal tourism resources and various recreational activities.Travel Cost Method (TCM) and Contingent Valuation Method (CVM) are two primarily important approaches used in monetarily valuing non-market social public environmental resources, and also vitally supporting method of estimating monetary compensation for ecosystem. This research, respectively compares, analyzes and summarizes TCM and CVM with respect to both theoretical basis and empirical models. The study is mainly focused on how to estimate models'parameters and calculate consumer surplus of tourism resources and recreational activities using the estimated econometric models and relevant statistical estimation techniques. Meanwhile, each model's performance is evaluated, and then the capability and adaptability between models are compared. TCM is categorized as continuous function, BOX-COX, count and other probability distribution models such as normal distribution, and nested-Logit etc.; And CVM consists of continuous function, Logit, Probit, logistic double-bounded dichotomous choice models and other probability distribution models such as Weibull distribution. Induce together with deduce, and empiricism blended theorization, are basic research mode. Synthetic analysis of those models and approaches is used to derive the proved procedure of formulas against obscurely arranged formulas in reviewed literatures. And summarized content can be used for depth analysis of TCM or CVM models. Followed up the comparison and sublimation, an application of valuing coastal tourism resources in Dalian is carried out.In this study, there are four special focuses of attention.(1)There are few subclasses for coastal natural tourism resources in Chinese GB/T18972-2003 due to hardly defining. The study points out that it is meaningful to define coastal natural tourism resources through social public environmental resource scarcity. Practically, for potential tourists, the scarcity of tourism resources can distinguish from the diversity, particularity and attraction of tourism activities. Tourism resource involves use value and nonuse value, whereas economic value of tourism activity advanced with the times can be reflected through Willingness to Pay (WTP) for tourism environmental resources no matter when tourism resource is visited, thus to continually find and program carriers of tourism activities is a new prospect in defining coastal tourism resources.(2)In social science research,5-point or 7-point Likert scale is used to express the correspondent's behavior preference for market goods or services. While CVM, a widely used empirical method, is difficult to reveal the true preference information for the respondents, thus, empirical questionnaire with elicited technique must be adopted to obtain the correspondent's true preference. The method, such as iterating biding with complex response process in western countries, is a new empirical concept with theoretical and cognitive gap for native respondents, which causes difficulties in tourists'corporations during a survey. As for the preferred dichotomous choice type of survey questionnaire, it tends to create problems of starting bias and sampling size restriction. For decreasing starting bias and avoiding complicated empirical questions, a new modified three step procedure is proposed in this study, which is consisted of pretest, pilot test and formal survey design which uses a Top-down method to inquire the true preference of respondents. Doing so is also helpful in reduction of iterating biding trouble in the process of survey for double-bounded dichotomous choice model.(3)A finding is that non-parameter model of Bayesian MCMC regression can easily solve parameters of TCM's or CVM's models. These advantages and applications in economics and management are just started. In the study, MCMC sampling regression is first use to value coastal tourism resource, and a MCMC CVM model comparison with Probit is analyzed.(4)Analysis of econometric models includes two aspects including the model estimation and application. The grasp of MLE regression is inadequate in many literatures, such as the discrimination between z-test and t-test. In the study, program codes of some models are practiced using Matlab computer language, and the model analysis tools are armed with those programs, furthermore the recognition of model test is carried out carefully. Currently, due to none of software that can be directly applied to compute MLE of double-bounded dichotomous choice model, its program code has to be used in solving out the model parameters.