If I have a set of data of 9 predictor variables (2 are numerical and 7 are categorical) and one numerical response variable. How can I find the correlation between the different variables and what type of regression can I apply to this in R? My question might be vague, but I don;t know that much about statistics and I have to solve this for a research. I appreciate any help.

Thank you

  • 3
    $\begingroup$ The main point of your question is not about "how to do" but "what to do" in statistics. This question does not belong here from my point of view. Furthermore you are asking an approach to a problem without providing a minimum effort to find out by yourself at least what kind of informations might be useful for you. $\endgroup$
    – SabDeM
    Jun 10 '15 at 16:31
  • 2
    $\begingroup$ Can you say some more about why you are doing the analysis? The size of your data set (number of observations, number of levels for each categorical variable)? The characteristics of your response variable (e.g. might it be reasonable to assume that the values conditioned on a particular set of predictors would be approximately Normal?) As a start, a general linear model (lm(y~.,data=mydata)) might be reasonable ... $\endgroup$
    – Ben Bolker
    Jun 10 '15 at 16:51

Your Answer

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

Browse other questions tagged or ask your own question.