Multiple regression - minitab
I am getting back to you regarding the multiple regression analysis. I have found your suggested data set on the statistical website:
and decided to calculate the body fat (so I am skipping the density - for my understanding - it is not for my project).
I have included the attachment with minitab.
1) is it correct?
2) When you look at the ind explanatory variables, it looks that only
have p-values less than .05 alpha
3) what represent a high riks or a fact of multicorreality ( i will fo the VIF test, and correlation coefficicient).
and when you look individually at t test for eac hvariable it looks that only:
wrist have value > than 1.96 (t/2 test for 5/2 alpha = .025).
So I should reduce the model but I am not sure what is the pefect one.
There are some unusual observations as well - I thought I could check them )Cook's values, DFits, H-leverage later)
Could you please send me the minitab attachment with the perfect model
and explain your decision?
I have a very detailed paper with lots of calculations, and I am planning to do everything (test for fits, multicorreality, stepwise, etc later). I would like to send you all my cal later and check and help me to correct mistakes.
I have 91% for A now, but still have a comprehensive exam ehead of me and this project. And I am actually not good in this subject, I really need help to excell here.
I would be very happy if you could help me to do it well.
For now, could you send me the
and scatterplot of the best linear line to make sure it is ok.
Also could you help me to explain which variables I should eliminate.
This question has the following supporting file(s):
This answer includes:
- Plain text
- Cited sources when necessary
- Attached file(s)
- Reduced model.doc