【主讲】Peihua Qiu from (Professor ，Department of Biostatistics in University of Florida)
【主题】Dynamic Screening System: An Approach For Early Detecting Irregularities of a Longitudinal Process
【时间】2018年6月8日 (周五) 15：30-17：00
【地点】上海财经大学经济学院楼401室
【语言】英文
【摘要】Big data often take the form of data streams with observations of an underlying longitudinal process collected sequentially over time. The performance of the process usually changes with time even when the process is considered stable or in-control. Such a process is called dynamic process in the literature. One critically important problem in studying the behavior of a dynamic process is to monitor its observations sequentially, so that any irregular behaviors can be detected as soon as possible. This dynamic screening problem has broad applications in engineering, public health, environmental protection and many other areas and disciplines. It should have many applications in economic research as well. The research team of the speaker developed a new and effective methodology recently, called dynamic screening system (DySS), for solving the dynamic screening problem. This method combines the strengths of statistical process control and longitudinal data analysis techniques. It is shown to be effective for many dynamic screening applications.
