Influence Analysis on the Direction of Optimal Response

• Main Authors: Jack Wang, 王聖文
• Other Authors: 黃郁芬
• Format: Others
• Language: en_US
• Published: 2011
• Online Access: http://ndltd.ncl.edu.tw/handle/55864570627892293721

碩士 === 國立中正大學 === 數理統計研究所 === 99 === In statistics, regression model and design of experiments are two useful tools for data analysis. Response surface methodology consists these two elements, and searches for the optimal path in the response surface which increases (or decreases) fast to save the cost of experiments or to improve the quality of products. Canonical analysis is a common method to find the direction of optimal response. Once an unusual point exists in the data, it is essential to detect this point which affects the optimal path seriously. The perturbation theory provides a useful tool in sensitivity analysis. In this thesis, we develop the influence functions via perturbation scheme to detect the unusual observations for the direction of optimal response. Some examples are provided to illustrate the applications of these methods.