Alternative Empirical Approaches to Correcting On-site Sampling Bias with an Application to Recreation Demand Analysis
2016/05/24
研討會日期 : 2016-05-24
時間 : 14:30
主講人 : Professor Ju-Chin Huang
主持人 : Professor Daigee Shaw
地點 : Conference Room B110
演講者簡介 : Professor Huang received her Ph.D. in Economics and Statistics at North Carolina State University in 1994. She is currently a Professor at University of New Hampshire. Her research fields are Environmental Valuation and Applied Econometrics.
演講摘要 : Collecting data via on-site surveys is convenient and can be cost effective. However, the on-site sampling scheme oversamples frequent site visitors and omits non-visitors, which can result in biased and inconsistent estimation of population parameters. In principle, to address the sampling issues embedded in on-site survey data, one may rescale the population distribution to account for the endogenous stratification and the truncated distribution. Econometric models for on-site data analysis based on specific distributional assumptions were developed (e.g., Shaw, 1988 and Englin and Shonkwiler, 1995). The shortcoming of these standard models is that the correction for on-site sampling bias is only accurate when the assumed distribution is correct. In this study, we propose two alternative empirical approaches that both utilize the sample distribution and treat the endogenous stratification and truncation issues separately. Monte Carlo simulation is conducted to evaluate alternative empirical approaches to correcting on-site sampling bias. The results show that the proposed empirical approaches outperform when the population distribution deviates from the assumed distributions in the standard models. A case study of the recreation demand for coastal beaches on Plum Island, Massachusetts is presented using the alternative correction methods.
