Creating a loop that will run a simple regression across all Independent variables

Question

I have built a simple regression model using the iris data set. I would like to run the dependent variable (iris$Sepal.Length) with all available variables, each regression with its own variable (each regression should be with . The outcome that I wanted to get is a list of each regression summary. Using the iris data set there should be 4 regressions:

r1 <- lm(Sepal.Length ~ Sepal.Width, data=iris)
r2 <- lm(Sepal.Length ~ Petal.Length, data=iris)
r3 <- lm(Sepal.Length ~ Petal.Width, data=iris)
r4 <- lm(Sepal.Length ~ Species, data=iris)

and a summary with the coefficients for each one of the regressions. Any Ideas how can I do this?

score 2 · Answer 1 · answered Mar 09 '14 at 09:32

2

library(plyr)
coefs = llply(names(iris)[-1], function(x){
  fml = as.formula(sprintf("Sepal.Length ~ %s", x))
  coef(lm(fml, data = iris))
})

answered Mar 09 '14 at 09:32

Ramnath

54,439
16
125
152

score 2 · Accepted Answer · answered Mar 09 '14 at 09:33

2

sapply(names(iris)[-1], 
       function(x) lm.fit(cbind(1, iris[,x]), iris[,"Sepal.Length"])$coef)
#   Sepal.Width Petal.Length Petal.Width  Species
#x1   6.5262226    4.3066034   4.7776294 4.261333
#x2  -0.2233611    0.4089223   0.8885803 0.791000

answered Mar 09 '14 at 09:33

Roland

127,288
10
191
288

Creating a loop that will run a simple regression across all Independent variables

2 Answers2

Linked