Loop over data table columns and apply glm using for loop

Question

I am trying to loop over my data table columns and apply glm to each column using a for loop.

for(n in 1:ncol(dt)){
  model = glm(y ~ dt[, n], family=binomial(link="logit"))
}

Why doesn't this work? I am getting this error:

Error in `[.data.table`(dt, , n) : 
  j (the 2nd argument inside [...]) is a single symbol but column name 'n' is not found. Perhaps you intended DT[, ..n]. This difference to data.frame is deliberate and explained in FAQ 1.1.

I nearly managed to make it work using something like dt[[n]], but I think it gets rid of the column name.

Ronak Shah · Accepted Answer · 2021-03-30 04:54:28Z

3

Using lapply to iterate over columns and reformulate to construct the formula.

model_list <- lapply(names(dt), function(x) 
                     glm(reformulate(x, 'y'), dt, family=binomial(link="logit")))

answered Mar 30, 2021 at 4:54

Ronak Shah

391k20 gold badges173 silver badges237 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

akrun · Accepted Answer · 2021-03-30 17:28:09Z

0

We can create a formula with paste and use that in glm

model <- vector('list', ncol(dt))
for(n in 1:ncol(dt)){
     model[[n]] = glm(as.formula(paste0('y ~ ',  names(dt)[n])),
               data = dt, family=binomial(link="logit"))
     }

edited Mar 30, 2021 at 17:28

answered Mar 30, 2021 at 0:59

akrun

891k38 gold badges590 silver badges700 bronze badges

2 Comments

user572780 Over a year ago

This doesn't work for me. I get a ` Error in eval(predvars, data, env) : object 'y' not found `

akrun Over a year ago

@user572780 I guess this is the similar question you asked there. It works with as.formula

Collectives™ on Stack Overflow

Loop over data table columns and apply glm using for loop

2 Answers 2

Comments

2 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

2 Comments

Your Answer

Sign up or log in

Post as a guest

Related