頁籤選單縮合
題名 | 二項配對數據之分析=Analysis of Matched Pair Dichotomous Data |
---|---|
作者 | 王堯弘; 林正祥; | 書刊名 | 中國統計學報 |
卷期 | 34:2 1996.06[民85.06] |
頁次 | 頁214-237 |
分類號 | 319.5 |
關鍵詞 | 配對; 二項數據; McNemar檢定; 卡方檢定; 混合卜瓦松逼近; Matched pairs; Dichotomous data; McNemar's test; Chi-square; Non-parametric test; Compound poisson approximation; |
語文 | 中文(Chinese) |
中文摘要 | 通常,欲檢定配對之二項分佈資料,皆以McNemar檢定行之。本文將就此作深入探討並指出McNemar檢定之缺點,提供另一統計量以代之,從而導出該統計檢定量之確實分佈及漸近分佈,並比較其和McNemar檢定之檢力。 |
英文摘要 | Given a random sample of n matched pairs dichotomous observations (X1, Y1), …,(Xn, Yn), each falling into one of these four cells: (0, 0), (0,1), (1,0), (1, 1), to test the null hyothesis Ho: P(X=l)=P(Y=l), the existing procedure is called McNemar’s test. This article reviews and criticizes McNemar’s test and proposes a new non-parametric procedure as an alternative. The exact and asymoptotic distributions of the test statistic are derived. Powers of the new and McNemar's tests are compared. |
本系統之摘要資訊系依該期刊論文摘要之資訊為主。