Contents
How do you perform a dominance analysis?
Dominance analysis (DA) is a method used to compare the relative importance of predictors in multiple regression. DA determines the dominance of one predictor over another by comparing their additional R2 contributions across all subset mod- els.
What is dominant predictor?
DA is utilized to determine if a predictor variable is “dominant” over another predictor. That is, that the dominant predictor is more related to the dependent variable both in pairwise predictor comparisons as well as in the presence of all possible combinations of predictors.
What are some of the measures used to assess the relative importance of predictors in multiple regression?
The analysis suggests that the General Dominance Index and Johnson’s Relative Weight are preferred methods for quantifying the relative importance of predictors in a multiple linear regression model.
What is relative importance analysis?
Relative importance or relative weight analysis is a method to “partition explained variance among multiple predictors to better understand the role played by each predictor in a regression equation” (Tonidandel & LeBreton, 2011).
How is dominance analysis used to compare predictors?
Dominance analysis provides an intuitive understanding of the contribution of a predictor because it establishes relative importance based on changes in R 2 as a function of adding predictors of interest to a model (Azen & Budescu, 2003; Nimon & Oswald, 2013). The method considers each pair of predictors (e.g.
How is the dominance library used in PCA?
The library can be used in combination with Principal Component Analysis (PCA) or Factor Analysis or any other feature reduction algorithm for getting accurate and intutive importance of predictors.
How are dominance weights used in statistical analysis?
Dominance weights were estimated over 1000 bootstrap samples to provide a means for a statistical test of differences in the relative importance of predictors (Azen & Budescu, 2003).
When to use a covariance matrix in dominance analysis?
Allows performing Dominance Analysis even in the cases where only the Covariance / Correlation matrix of the predictor variables is available. Provides the user the flexibility to choose number of top predictors that they want to compute relative importance for.