| Type: | Package |
| Title: | Shapley Value Regression for Relative Importance of Attributes |
| Version: | 0.2.0 |
| Author: | Jingyi Liang |
| Maintainer: | Jingyi Liang <jingyiliang19@163.com> |
| Description: | Shapley Value Regression for calculating the relative importance of independent variables in linear regression with avoiding the collinearity. |
| License: | MIT + file LICENSE |
| Encoding: | UTF-8 |
| RoxygenNote: | 7.1.1 |
| Suggests: | knitr, rmarkdown |
| Imports: | tidyverse,kableExtra,MASS,utils |
| VignetteBuilder: | knitr |
| NeedsCompilation: | no |
| Packaged: | 2021-07-27 02:54:36 UTC; liji0004 |
| Repository: | CRAN |
| Date/Publication: | 2021-07-27 04:50:01 UTC |
ShapleyValueRegression – to calculate the relative importance of attributes in linear regression
Description
Shapley Value Regression for calculating the relative importance of independent variables in linear regression with avoiding the collinearity.
Arguments
y A coloumn or data set of the dependent variable
x A matrix or data set of the independent variables
Value
The structure of the output is a datatable, with two rows:the unstandardized and standardized relative importance of each attributes using shapley value regression method.
Examples
library(MASS)
library(tidyverse)
data <- Boston
y <- data$medv
x <- as.data.frame(data[,5:8])
shapleyvalue(y,x)