I am currently writing my MSc thesis and I am using a Tobit model.

My data consists of fleet and sales data. The censored dependent variable is quantity and this quantity is higher than the fleet number. However, this fleet number differs for every line.

I found that I could use the crch package in R and use the fleet numbers as a right censoring point. Or I think I can use that.

But I am wondering, is there also another option to estimate the regression with a changing number as censoring point?

Thanks in advance!


Yes, you can have that each sample unit in your dataset has a different censoring point. You can easily do that using the survreg() function from the survival package in R. Using the Surv() function you can define the type of censoring you have in the formula argument of survreg(), and with the dist argument you can define the distribution of your outcome, e.g., "gaussian".

| cite | improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.