1
$\begingroup$

I am currently researching fish survival analysis. The weight and number of live fish were observed until the end of the rearing period, and the number, weight, and timing of dead fish during the rearing period were observed. Accordingly, I wanted to create a fish survival model using the weight of the fish as a covariate.

The weight,(time-varying covariate) was observed only once, at death/censoring time. Is there a way to deal with time-varying covariates that are observed only once for each individual? If not, should we create a survival model assuming it is a time-independent covariate? I wonder if there is another solution.

Improve this question
$\endgroup$
1
  • 1
    $\begingroup$ What is your objective here? That is, what are you trying to estimate with your model? $\endgroup$
    – George Savva
    Commented Nov 30, 2023 at 11:04

1 Answer 1

Reset to default
0
$\begingroup$

Well since you measured it only once it is not a time varying covariate. It is something that varies with time yes, but in your case it is not. So indeed you have to treat it as time-independent.

Improve this answer
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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