비모수 접근법에 의한 호텔객실 조망권 가치평가

The purpose of this study is to examine a nonparametric estimator suitable for discrete response valuation experiments. There are two significant advantages of the approach over available parametric methods; first, no distribution assumption is necessary and, second, it is very simple to use. Another advantage of the nonparametric estimator is that its variance is readily calculated under the assumption of independence of the responses. The parametric and non-parametric approaches are applied to the positive externalities provided by ocean, mountain, and city as a scenic landscape viewed from a hotel room, respectively. The nonparametric estimates are compared with the estimated truncated means from the logit model. The standard errors of the nonparametric estimators are used to construct confidence intervals around benefit estimates to determine whether there are significant differences in willingness to pay(WTP) estimates between the two approaches and among the three views. For the data used, the estimates of the parametric logit and nonparametric approaches are in fairly close agreement and show the same ordering for the non-market products. The confidence intervals around the mean WTP's derived from the nonparametric estimators also result that there is a significant difference in the benefits among the views.