Relative Importance for Linear Regression in R: The Package relaimpo

By U. Groemping. 2006

2,651 citations recorded by the source corpus.

Relative importance is a topic that has seen a lot of interest in recent years, particularly in applied work. The R package relaimpo implements six different metrics for assessing relative importance of regressors in the linear model, two of which are recommended - averaging over orderings of regressors and a newly proposed metric (Feldman 2005) called pmvd. Apart from delivering the metrics themselves, relaimpo also provides (exploratory) bootstrap confidence intervals. This paper offers a brief tutorial introduction to the package. The methods and relaimpo's functionality are illustrated using the data set swiss that is generally available in R. The paper targets readers who have a basic understanding of multiple linear regression. For the background of more advanced aspects, references are provided.

Subfields: Artificial Intelligence, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/1ce789bc60ba0a556c5ae848dde03a0b0e74384a

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Statistics and Probability

Citations in the corpus

Cross-scale variation in species richness–environment associations (2011, methodology): “We therefore used hierarchical partitioning to assess the average contribution of each predictor to the variance in species richness over all possible models (Grömping, 2006; Murray & Conner, 2009).”
Axiomatic arguments for decomposing goodness of fit according to Shapley and Owen values (2012, background): “For R(2) as GOF, this burden can be alleviated to some extent if the calculation is based on the covariance structure of the variables rather than the individual observations [6].”
Peer feedback on academic writing: undergraduate students’ peer feedback role, peer feedback perceptions and essay performance (2018, methodology): “1, R Core team 2017), the ‘relaimpo’ package (Groemping 2006) was applied to compare the relative contribution of the independent variables.”
Random forest as a generic framework for predictive modeling of spatial and spatio-temporal variables (2018, methodology): “Decomposition of R(2) as often used for linear models (Groemping, 2006) neglects model 587 selection and does not straightforwardly apply to kriging.”
Variation in seed dispersal effectiveness: the redundancy of consequences in diversified tropical frugivore assemblages (2016, neutral): “We then performed a multiple regression of each subcomponent on QC (lmg estimates, package relaimpo, Grömping 2006 in R package ver.”
Variable Importance Assessment in Regression: Linear Regression versus Random Forest (2009, background): “those in Grömping (2007); for these scenarios, LMG and”
Understanding the Results of Multiple Linear Regression (2013, background): “Recently, researchers have developed and provided computer programs, macros, and software packages to compute the metrics in linear regression to be discussed here, such as dominance analysis (Azen & Budescu, 2003) and relative weights (Grömping, 2006; Tonidandel, LeBreton, & Johnson, 2009). Althoug”
Mental Disorders as Causal Systems (2015, methodology): “To calculate relative importance, we used the lmg metric in the R package relaimpo (Grömping, 2006).”
Working on my own: Measuring the challenges of gig work (2021, methodology): “We also ran relative weights analysis (Johnson, 2000) to assess the relative contribution of each gig challenge subfactor on the outcome of depletion using the relaimpo package in R (Grömping, 2006).”