For the data analysis of a clinico-epidemiological study, I intend to select four seperate dependent variables to assess the correlation with independent variables. These all the independent variables are categorical (Nominal) variables consist of two or more than two categories. Three dependent variables out of four consist with more than three categories. According to my knowledge I wish, Multinominal logistic regression would be the most appropriate regression type. Since other dependent variables has only two categories, I prefer to go head with binomial logistic regression analysis. I you could share your thoughts, I would appreciate and be pleased.
Thank you..
I agree with your assessment. If prediction is your goal you may find the attachment material may be of some interest. Best wishes David Booth
As you have more than one dependent Variable, multinomial logistic regression may be the suitable to the data.
I highly appreciate all of your valuable comments David Eugene Booth Medhat Elsahookie Abdulmuhsin S. Shihab
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Regression