Importance of vars in linear regression (R)

[juanvg1972]

I have a question that is and R question and a statistical question:
I am analysing sales of a retailer. These sales are related to some vars: var1, var2, var3.., varN
Most of the vars are continuos.
I want to analyze the relationship between sales and the vars. I have made a linear regression with R:

rg<-lm(sales ~ var1 + var2 + var3 + var4, data=sales_2017)
summary(rg)

Now I want to know which is the most important variable in sales, and to know the percent of importance of each var. I am doing this (caret package):

varImp(rg, scale = FALSE)
rsimp <- varImp(rg, scale = FALSE)
plot(rsimp)

Is this a good method to obtain variables importance??, is good way in R?
Thanks in advance. Any advice will be greatly apreciated.

Juan

 

[hlsmith]

What does the scale =false option mean here?

 

[noetsi]

There is no easy/agreed on way to determine the relative impact of a variable in regression. I spent years looking in the literature and asking people, including here, to get at this and my conclusion is that regression was really not designed to answer this question surprising as that is to me.

In logistic regression the strength of the wald statistic can be used to rate relative impact. I forgot what I used for linear regression, but I will look it up. But again there is no agreement of what the best way to do this - and most discussions of regression do not address it including books on regression.