How can I use tryCatch to count errors in a while try loop

Question

I want to be able to count my error/non-convergence from my lm model, which the try function skips, using a tryCatch, I will give an example of my code below so someone can help.

i =- 0
count = 0
count1 = 0
fit = rep(NA, 1000)

while(i < 1000) try({ 
    count1 = count1 + 1
    dat = sample(data,replace = T)
    fit[i] = lm(y~x1 + x2, data = dat)
    count <- count + 1
}, silent = T)

I have this code above that runs smoothly but sometimes because I resample the data, model does not converge and hence the use of try function, and I need to count the number of times that the model doesnt converge.

I tried putting count1 and count before and after the model and they are all counting 1000. But I know for sure that some of the models don't converge and give errors (thus if x1 categorical is such that it has only one level instead of two levels after resampling then we have errors).

Can someone help me how I can use tryCatch inside this while loop to be able to solve this problem?

Thomas · Accepted Answer · 2013-09-03 12:08:01Z

1

You need something like this (and you need to increment your while loop):

i <- 1
count1 <- 0
fit <- rep(NA, 1000)

while(i <= 1000) { 
    dat <- sample(data, replace = T)
    temp <- try(lm(y~x1 + x2, data = dat))
    if(inherits(temp,"try-error"))
        count1 <- count1 + 1
    else
        fit[i] <- temp
    i <- i + 1
}

answered Sep 3, 2013 at 12:08

Thomas

44.7k12 gold badges115 silver badges144 bronze badges

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

Collectives™ on Stack Overflow

How can I use tryCatch to count errors in a while try loop

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related