Reputation: 4640
I have a dataframe with many columns, and I want to do a regression.
Currently, I have to type:
lm (col3 ~ col4 + col5 + col6 ... + col20)
It is not very convenient. Is there a faster way to do that?
Upvotes: 1
Views: 33
Reputation: 887148
We subset the dataset by selecting the columns that are needed in the regression formula and use .
on the RHS of ~
to specify all the columns that are not the dependent variable ('col3').
lm(col3~., subset(df1, select=col3:col20))
set.seed(24)
df1 <- as.data.frame(matrix(sample(0:9, 25*40, replace=TRUE), ncol=25))
colnames(df1) <- paste0('col', 1:25)
Upvotes: 3