dominance analysis是用来检测各个自变量之间的相对重要性,在stata中help domin就可以使用。根据stata给出的例子,可以应用于很多模型,如:
Example 1: linear regression dominance analysis
Example 2: Ordered outcome dominance analysis with covariate (e.g., Luchman, 2014)
Example 3: Binary outcome dominance analysis with factor varaible (e.g., Azen & Traxel, 2009)
Example 4: Comparison of interaction and non-linear variables
Example 5: Epsilon-based linear regression approach to dominance with bootstrapped standard errors
Example 6: Multivariate regression with wrapper mvdom; using default Rxy metric (e.g., Azen & Budescu, 2006; LeBreton & Tonidandel,2008)
Example 7: Gamma regression with deviance fitstat and constant-only comparison using reverse
Example 8: Mixed effects regression with wrapper mixdom (e.g., Luo & Azen, 2013)
Example 9: Multinomial logistic regression with simple program to return BIC
Example 10: Multiply imputed dominance analysis
最近我在使用的过程中,也遇到了问题,分类变量是不是不能用于dominance analysis?希望有了解的同仁能够赐教,多谢!
|