1
$\begingroup$

I have independent continuous variables like Age, Price and outcome variable like purchased or not purchased

I know if we have continuous variables on input and output, we can use pearson correlation or chi-square test if both input and output are categorical.

Now, which test should I use to test the association of continuous input variable with categorical outcome variable?

I already referred this post but there is no response.

So, can someone please help me?

$\endgroup$
3
$\begingroup$

With a binary outcome just use regression. There are already very many posts at this site, so search for more information.

If you want more specific advice, you need to update (edit) your post with specific information.

$\endgroup$
6
  • $\begingroup$ I have only two input continuous variable. Should i still use log reg? Can't i individually test the association ofone inputvariable to the outcome variable? Can log reg work with1 input variable and 1 outputvariable? ..upvoted $\endgroup$ – The Great May 25 at 23:54
  • $\begingroup$ halvorsen - can't we use point biseral coorelation? $\endgroup$ – The Great May 26 at 0:07
  • $\begingroup$ Well, maybe, it depends on your goal. Why do you think logistic regression is not a good soultion? $\endgroup$ – kjetil b halvorsen May 26 at 18:28
  • $\begingroup$ Am happy to use logistic regression. But due to my lack of experience, i don't know whether logistic regression can be applied to estimate the relationship between one input variable amd one binary output variable? $\endgroup$ – The Great May 27 at 2:42
  • $\begingroup$ If logistic regression can be used for one input and one binary output variable, may I know what is the use of point biseral correlation then? When to use point biseral correlation? How is it dofferent from logostic regression? $\endgroup$ – The Great May 27 at 2:43

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.