交叉設計模型的影響分析

Influence Analysis on Crossover Design Model

柯博祥(Ke,Boshiang)

指導教授 ： 黃郁芬

國立中正大學/理學院/數學系統計科學研究所/碩士(2012年)

摘要

在生體相等性研究上，交叉設計模型是最普遍的統計模型。此研究的評估通常是根據受試者間的變異。然而，此研究的結果容易受到影響受試者和異常值得影響。Chow 和 Tze 在1990年討論生體相等性研究的這個議題並提出概似距離和估計距離來偵測可能的異常值。在此篇論文，我們發展影響函數使其成為另外一種方法來偵測影響受試者。此外，影響函數也有與Chow 和 Tze提出的方法，亦有透過一組模擬資料和兩組實際資料來加以比較。

關鍵字

交叉設計模型 ； 影響函數 ； 受試者內變異

並列摘要

Crossover design model is a commom statistical model for the bioequivalence study. The evaluation of this study is usually based on the intra-subject variability. However, the result of this analysis can be affected by influential subjects or outliers. Chow and Tze (1990) discussed this issue based on the likelihood distance and estimates distance to detect the potential outliers for the bioequivalence study.
In this thesis, we develop the influence functions as an alternative approach for detection of influential subjects. Moreover, the comparison between the proposed approach and the methods proposed by Chow and Tze are investigated. One simulated data and two real data are provided to illustrate the results of these approaches.

並列關鍵字

crossover design model ； influence functions ； intra-subject variability

參考文獻

1. Chow, S. C. and Tse, S. K.(1990). Outlier detection in bioavailability/bioequivalence studies. Statistics in

Medicine, 9, 549-558.

2. Chow, S. C. and Liu, J. P. (2008). Design and Analysis of Bioavailability and Bioequivalence Studies, 3rd

3. Clayton, D. and Leslie, A. (1981). The bioavailability of erythromycin stearate versus enteric-coated

erythromycin based when taken immediately before and after food. Journal of International Medical Research,