dominanceAnalysis

Dominance Analysis (Azen and Bodescu, 2003, 2006; Azen and Traxel, 2009;Luo and Azen, 2013), for multiple regression models: OLS, GLM and HLM.

Features:

• Provides complete, conditional and general dominance analysis for lm, lmer, glm(family=binomial) and correlation matrix based regression
• You could use multiple criteria as fit indexes (useful for HLM)

With lm (see Azen and Bodescu, 2003)

dominanceAnalysis(lm(y~x1+x2+x3))

For hierarchinal linear models using lmer, you should provide a null model. Requires https://github.com/clbustos/r-glmmextra (see Luo and Azen, 2013).

dominanceAnalysis(lmer(y~x1+x2+x3+(1|g)), null.model=lmer(y~(1|g)))

For logistic regression (experimental) (see Azen and Traxel, 2009).

dominanceAnalysis(glm(y~x1+x2+xx3,family=binomial))

Using correlation matrix based on a data.frame

lwith<-lmWithCov(y~x1+x2+x3+x4,cor(d.f))
dominanceAnalysis(lwith)

Bootstrap analysis (Azen and Bodescu, 2006)

bootDominanceAnalysis(lm(y~x1+x2+x3))

Contributors

• Filipa Soares: Bug report.

References

• Azen, R. & Budescu, D. (2003). The Dominance Analysis Approach for Comparing Predictors in Multiple Regression. Psychological Methods, 8 (2), 129-148.
• Azen, R. & Budescu, D. (2006). Comparing predictors in Multivariate Regression Models: an extension of Dominance Analysis, 31(2), 157-180.
• Azen, R. & Traxel, N. (2009). Using Dominance Analysis to Determine Predictor Importance in Logistic Regression. Journal of Educational and Behavioral Statistics,34(3), 319-347.
• Luo, W. & Azen, R. (2013). Determining Predictor Importance in Hierarchical Linear Models Using Dominance Analysis. Journal of Educational and Behavioral Statistics, 38(1), 3-31.