多組數效用量估計

Abstract

多組數效用量，即標準化期望值之變異數與標準差，是用以表示多組不同處理之間其總變異程度的指標，對於統計結果有實質性的解釋；因此在研究之中加入多組數效用量一向被視為是良好的作法。本研究中使用SAS程式之IML語法進行模擬，在非中心F分配之變異數分析的固定效果模式下，模擬不同點估計量在不同的參數組合下其表現與差異，試圖找出最佳的標準化期望值之變異數與標準差之點估計量。

Effect size which is used to estimate the magnitude of effect or association between two or more treatments is a common statistical approach, and it is a very useful tool for researchers to interpret their results. This paper uses SAS/IML program to generate some random samples to compare three types of point estimators of effect size. We try to find out the most appropriate point estimator for different situation, and hope to provide some contribution to other researchers.