我需要邏輯斯迴歸樣本數計算的標準或公式都可以,
但是一定要有文獻出處!!
將邏輯斯迴歸方程式寫成:
圖片參考:http://img501.imageshack.us/img501/968/logicnlu1.p...
文獻出處:Hsieh, F.Y., Block, D.A., and Larsen, M.D. (1998). A Simple Method of Sample Size Calculation for Linear and Logistic Regression. Statistics in Medicine, Volume 17, pages 1623-1634.
如果怕手算算錯,也可以利用網路資源(Sample Size Calculation for Logistic Regression),直接計算所需樣本數:http://www.dartmouth.edu/~eugened/power-samplesize...
2008-06-30 11:48:42 補充:
另外,根據「Harrell, F. E. (1984). "Regression modelling strategies for improved
prognostic prediction." Statistics in Medicine 3: 143-152.」,邏輯斯迴歸中每一個自變項至少要有十個cases。如果研究中有1萬個病患,而其中100人死亡,換言之,研究中有100個cases,這時候邏輯斯迴歸的自變項不應超過十個。
2008-07-02 08:31:30 補充:
也請參考鄭光甫教授,「分析logistic 迴歸模型中所必需的最小樣本數」:http://web2.go2upload.com/file/24185
Copyright © 2024 1QUIZZ.COM - All rights reserved.
Answers & Comments
Verified answer
將邏輯斯迴歸方程式寫成:
圖片參考:http://img501.imageshack.us/img501/968/logicnlu1.p...
文獻出處:Hsieh, F.Y., Block, D.A., and Larsen, M.D. (1998). A Simple Method of Sample Size Calculation for Linear and Logistic Regression. Statistics in Medicine, Volume 17, pages 1623-1634.
如果怕手算算錯,也可以利用網路資源(Sample Size Calculation for Logistic Regression),直接計算所需樣本數:http://www.dartmouth.edu/~eugened/power-samplesize...
2008-06-30 11:48:42 補充:
另外,根據「Harrell, F. E. (1984). "Regression modelling strategies for improved
prognostic prediction." Statistics in Medicine 3: 143-152.」,邏輯斯迴歸中每一個自變項至少要有十個cases。如果研究中有1萬個病患,而其中100人死亡,換言之,研究中有100個cases,這時候邏輯斯迴歸的自變項不應超過十個。
2008-07-02 08:31:30 補充:
也請參考鄭光甫教授,「分析logistic 迴歸模型中所必需的最小樣本數」:http://web2.go2upload.com/file/24185