時 間：97年7月8日(星期二) 15:00 – 16:00
演講人：HUNG T. NGUYEN (New Mexico State University, USA.)
題 目：STATISTICAL ANALYSIS OF COARSE DATA AND RELATED RESEARCH PROBLEMS
綱要：Applied statistical research problems arise often from the lack of precision in the observed data. This is illustrated by well-known situations such as measurement-error in linear regression, censored data in survival analysis, missing data in econometrics, hidden data in biology, indirect observed data in auctions, interval-valued statistics, and more generally coarse data in which the range of a variable of interest is replaced by a coarsening so that the desirable data are hidden inside. These are imprecise data in different forms. Standard statistical procedures for estimation, hypothesis testing and forecasting cannot be directly applied with such data. Thus, prior to carry out statistical analysis of real-world problems, it is necessary to modify appropriately standard procedures to best cope with imprecise data to make the applications meaningful. Clearly, this has to be done on a case-by-case basis, i.e. for each different type of imprecise data. Many recent efforts are devoted to developing statistical tools for dealing with inferences from a wide rand of data, which may be structured in many different ways. In this talk, we will illustrate several typical research issues to bring out methodologies and techniques for research in applied statistics.
地 點：果夫樓研討室二