Multiple linear regressions

TopQ

New member
Joined
Mar 11, 2016
Messages
1
Hello All
I have problem trying to do multiple linear regression to test one single variable against many other variables.
This is the problem:
I have one numerical variable (a) and I want to see if there is any correlation to other variables but I don't know how to do in a batch mode.
I know how to do a one vs one variable and then analyze the data but how do I perform the same operation when I have another 100 variables that I want to test against the original variable (a) I really don't have the time to test (a) vs all the other variable so is there an easier way to do that? What I want at the end is to be able to determine if any of the 100 + variables show any correlation with variable (a)

I do have access to Excel and Minitab as software packages.

Thanks for your help.
 
Hello All
I have problem trying to do multiple linear regression to test one single variable against many other variables.
This is the problem:
I have one numerical variable (a) and I want to see if there is any correlation to other variables but I don't know how to do in a batch mode.
I know how to do a one vs one variable and then analyze the data but how do I perform the same operation when I have another 100 variables that I want to test against the original variable (a) I really don't have the time to test (a) vs all the other variable so is there an easier way to do that? What I want at the end is to be able to determine if any of the 100 + variables show any correlation with variable (a)

I do have access to Excel and Minitab as software packages.

Thanks for your help.
Excel does multivariate analysis
https://www.google.com/search?q=mul...=utf-8&oe=utf-8#q=multivariate+analysis+excel
as does Minitab
https://www.google.com/search?q=mul...tf-8&oe=utf-8#q=multivariate+analysis+minitab
 
Hello All
I have problem trying to do multiple linear regression to test one single variable against many other variables.
This is the problem:
I have one numerical variable (a) and I want to see if there is any correlation to other variables but I don't know how to do in a batch mode.
I know how to do a one vs one variable and then analyze the data but how do I perform the same operation when I have another 100 variables that I want to test against the original variable (a) I really don't have the time to test (a) vs all the other variable so is there an easier way to do that? What I want at the end is to be able to determine if any of the 100 + variables show any correlation with variable (a)

I do have access to Excel and Minitab as software packages.

Thanks for your help.

Just to expand on the idea, theoretically it doesn't really doesn't make any difference whether the 'next dependence' is x12 [the same variable but squared] or x2 [a brand new variable]. That is whether the 'least squares fit' is to
y=a + b x1 + c x12
or to
y = a + b x1 + c x2.
 
Top