cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
JMP is taking Discovery online, April 16 and 18. Register today and join us for interactive sessions featuring popular presentation topics, networking, and discussions with the experts.
Choose Language Hide Translation Bar
nac
nac
Level III

Multivariate vs. Graph builder

Hi,

When I try to find correlation between Age and Height  by using multivariate analysis R2=0.9807

When I apply the same thing by using Graph Builder R2=0.962

 

What is the reason of this change. This difference will be increase based on results.

 

Could someone help on this?

 

You can also see jmp file on enclosed.

 

 

 

2 REPLIES 2

Re: Multivariate vs. Graph builder

I verified the Pearson linear correlation coefficient = 0.9807 from Multivariate with Bivariate and by hand. I do not know how the estimate provided by Graph Builder is calculated. You might send this example to JMP Technical Support (support@jmp.com) for investigation. Please report here what you learn.

Re: Multivariate vs. Graph builder

I think you are confusing the Pearson correlation coefficient (r) with the Coefficient of Determination (RSquare). 

From Fit Y by X, Multivariate, AND Graph Builder, I get the same value for r, which is -0.981. I have included the Graph Builder report that shows this in the upper-left of the graph.

GraphBuilder.JPG

 

If I fit a linear regression line to the data I get an RSquare of 0.962 from both Graph Builder and Fit Y by X.

FitYbyXRSquare.JPG

 

Graph Builder report for RSquare.

GBRSquare.JPG

Dan Obermiller