0
$\begingroup$

I have more than 100 variables and the outcome variable is continuous (percentage of remission). I was wondering what kind of model can I use for prediction of these types of response based on the fact that I cannot use classification due to continuous response outcome.

$\endgroup$

migration rejected from stackoverflow.com Aug 15 at 11:24

This question came from our site for professional and enthusiast programmers. Votes, comments, and answers are locked due to the question being closed here, but it may be eligible for editing and reopening on the site where it originated.

closed as too broad by kjetil b halvorsen, gung Aug 15 at 11:24

Please edit the question to limit it to a specific problem with enough detail to identify an adequate answer. Avoid asking multiple distinct questions at once. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.

  • 1
    $\begingroup$ Questions about stats methods and not programming are off-topic on SO but may be on-topic at Cross Validated. But this will probably still be too broad there: the "best" model is totally dependent on context, data, resources, etc. $\endgroup$ – camille Aug 13 at 16:08
  • $\begingroup$ Do you mean that your response variable is multivariate? $\endgroup$ – Dave Aug 13 at 16:47
  • 1
    $\begingroup$ This is probably too broad, & is in danger of being closed. You should see if you can make it narrower & more concrete. In the interim, what does "percentage of remission" mean, exactly? Is it, eg, the number of patients in remission out of a total number of patients treated? $\endgroup$ – gung Aug 13 at 16:52
1
$\begingroup$

I suggest you try the Generalised Linear Model with a Logit link function.

You may initially think a logistic regression only works on binary outcome variables, but that is not necessarily the case. Outcome variable which are proportions or percentages, such that they are continuous between $0$ and $1$, can also be predicted using the Generalised Linear Model with a Logit link function.

This may depend more on the software you use. I recommend statsmodels within Python. If you are using R, there are likely many similar packages.

$\endgroup$
1
$\begingroup$

Since you have many variables I would suggest Lassa or Elastic Net regression in oder to avoid oroblems with correlated predictors or overfitting. You can read more here

New contributor
AndroidApp is a new contributor to this site. Take care in asking for clarification, commenting, and answering. Check out our Code of Conduct.
$\endgroup$

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