In this week’s episode Greg and Patrick explore the extremely clever yet inexplicably underused method of dominance analysis which offers a set of techniques for determining the relative importance of predictors in a regression model. Along the way they also discuss giving compliments, looking tired, Indy vs. F1, chicken paprikás, Gustav Holst, Fozzie Bear, not paying attention while recording, Lewis Hamilton pin-ups, Lando Calrissian, equation forts, being appallingly cool, making no sense at all, and magnums of champagne.
Related Episodes
- S4E13: Model-Based Power Analysis: The Power of *What*
- S3E09: Semi-Partially Clarifying Measures of Association in Regression
- S2E18: Regression: Like That Old High School Friend You’ve Outgrown
- S2E11: The Replication…Dilemma with Samantha Anderson
Suggested Readings
Azen, R., & Budescu, D. V. (2003). The dominance analysis approach for comparing predictors in multiple regression. Psychological Methods, 8, 129.
Azen, R., & Traxel, N. (2009). Using dominance analysis to determine predictor importance in logistic regression. Journal of Educational and Behavioral Statistics, 34, 319-347.
Budescu, D. V. (1993). Dominance analysis: a new approach to the problem of relative importance of predictors in multiple regression. Psychological Bulletin, 114, 542.
Budescu, D. V., & Azen, R. (2004). Beyond global measures of relative importance: Some insights from dominance analysis. Organizational Research Methods, 7, 341-350.
Luo, W., & Azen, R. (2013). Determining predictor importance in hierarchical linear models using dominance analysis. Journal of Educational and Behavioral Statistics, 38, 3-31.
Mizumoto, A. (2023). Calculating the relative importance of multiple regression predictor variables using dominance analysis and random forests. Language Learning, 73, 161-196.