透過您的圖書館登入
IP:3.149.249.124
  • 學位論文

限制最大概似估計的應用

Using REML for the Mixed Effects Models.

指導教授 : 鄭子韋

摘要


在實務上,因子設計除了固定效果,還經常牽涉一個或數個隨機效果而形成混合效果模型。若研究者對模型中的變異成分感興趣,可根據變異數分析表,使用動差法估計來獲得,但其估計值缺乏最概估計量的優良統計性質,本文推薦的參數估計法為限制最大概似法。 本研究使用限制最大概似法來分析變異數成分,並探討其理論。再利用模擬混合效果模型,比較限制最大概似法、最大概似法、動差法,驗證在平衡資料下,限制最大概似法是否為較優良的方法。 最後,利用模擬資料來說明,如何使用不同的估計方法來分析變異數成分,並統整電腦軟體中語法的指定。

並列摘要


In practice, in addition to the fixed effects, the experimental design often involves one or several random effects to form a mixed effects model. Variance components of the model can be obtained from the ANOVA table by the method of moments (MME). However, MME doesn’t have the statistical property of the maximum likelihood estimator. The restricted maximum likelihood (REML) is another useful technique for analysing the mixed effects models. In this paper, we illustrate how to use REML to analyze the variance components, and compare REML with other methods.

參考文獻


[1] Arnold, S. F. (1981). The Theory of Linear Models and Multivariate Analysis, John Wiley & Sons, Inc.
[3] Bartlett, M. S. (1937). Properties of Sufficiency and Statistical Tests. Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences, 160(901), 268.
[4] Corbeil, R. R. and Searle, S. R. (1976).Restricted Maximum Likelihood (REML) Estimation of Variance Components in the Mixed Model. Journal of the American Statistical Association, 18, 31-38.
[5] Cressie, N. and Lahiri, S. N.(1993)The Asymptotic Distribution of REML Estimators. Journal of Multivariate Analysis, 45, 217-233.
[6] Drum, M. L. and McCullagh, P. (1993)REML Estimation with Exact Covariance in the Logistic Mixed Model Biometrics, 49, 677-689.

延伸閱讀