for an outcome (relapse=yes or no) i performed correlation which revealed 7 highly significant corelated variables, i tested then in binary logestic , the p value of all of them was 1.000 or 0.98>>>why the p value was not significant like that in the correlation
Great great video... Yes, I will be checking your other videos. It makes a lot more since and easier to select a statistical model when you intuitively understand the equations. I am a university professor and I see that my students do much better when they understand, not necessary use, the math behind the models. My question, you did not say anything about the "overall percentage correct", how does it interpreted? What does it tell me about the model? Would it be something similar to r2 (r squared) in simple regression?
Hi! First of all, thank you so much for the videos you posted they are handy material to refer to. Regarding this video, I would like to ask if you can make a follow-up video on ROC curve to decide the cutoff criteria with the probabilities to decide who are likely to be Republican or Democrat.
Very helpful, please keep it up!! I will also share your videos to my workmates. thank you!!
Can you please do multinominal logistic regression?!?!
for an outcome (relapse=yes or no) i performed correlation which revealed 7 highly significant corelated variables, i tested then in binary logestic , the p value of all of them was 1.000 or 0.98>>>why the p value was not significant like that in the correlation
Great great video... Yes, I will be checking your other videos. It makes a lot more since and easier to select a statistical model when you intuitively understand the equations. I am a university professor and I see that my students do much better when they understand, not necessary use, the math behind the models.
My question, you did not say anything about the "overall percentage correct", how does it interpreted? What does it tell me about the model? Would it be something similar to r2 (r squared) in simple regression?
Hi! First of all, thank you so much for the videos you posted they are handy material to refer to.
Regarding this video, I would like to ask if you can make a follow-up video on ROC curve to decide the cutoff criteria with the probabilities to decide who are likely to be Republican or Democrat.