Your browser does not support JavaScript!
2016-05-13(五),主講人:廖振鐸教授 (國立臺灣大學 農藝系)



統 計 學 研 究 所


專 題 演 講

講 題: Identification of Location and Dispersion Effects from Partially Replicated Two-level Factorial Designs
演講者: 廖振鐸教授 (國立臺灣大學 農藝系)
時 間: 105年05月13日(星期五)10:40 - 12:10  (上午10:20 - 10:40茶會於統計所821室舉行)
地 點: 綜合三館837室
摘 要: It has been emphasized on the variation reduction in industrial quality improvement. Thus, well designed experiments are usually required to identify important effects to control the quality characteristic of interest around a target value with a minimal variation. Several methods have been proposed to analyze both location and dispersion effects from unreplicated factorial designs. However, the use of unreplicated designs can be inherently difficult in the identification of the truly active effects since location and dispersion effects are possibly confounded. Between unreplicated and fully replicated two-level factorial designs, we promote compromised designs with partial replications. Based on the concept of generalized p-values, we develop a unified method to identify active location and dispersion effects. We first determine the significant dispersion effects from the repeated runs. Then, perform location effects identification by taking the identified active dispersion effects into account. The proposed procedure is illustrated with a real dataset, and evaluated through a detailed simulation study. The results support partially replicated designs for practical use. In addition, we will discuss the construction of a class of partially replicated designs.

敬請公佈                                 歡迎參加


  • 友善列印
  • 分享
Voice Play