Linear regression Equation?

#1
Hi All,

I need to determine whether two continuous variables are positively associated or not. I believe that this required a simple linear regression (?). But i'm not sure if my formula is correct:

Y= α + ß1

Where ß1 is one of my continuous variables and Y is the other?

Also, I don't think it really matters which variable is the independent or dependent. Does this affect my equation?
 

Dragan

Super Moderator
#2
Hi All,

But i'm not sure if my formula is correct:

Y= α + ß1

Where ß1 is one of my continuous variables and Y is the other?
(1) Your equation is not a common (or standard) way to express a simple linear regression model.

(2) The Pearson correlation has the property of symmetry. That is, the correlation between (two variables) X and Y is the same as the correlation between Y and X.
 

Dragan

Super Moderator
#4
Thank you...

What is wrong with my equation? Should it be instead:

Y= α + ß1* random variable ?
Yes, call it X, and you'll also need an error term because you're expressing the model with the actual values of Y and not the predicted values of Y.