Get Inspire insights from former attendees in our AMA discussion thread on Inspire Buzz. ACEs and other community members are on call all week to answer!

Alteryx Designer Desktop Discussions

Find answers, ask questions, and share expertise about Alteryx Designer Desktop and Intelligence Suite.
SOLVED

Regression model statistics

JuhaP
7 - Meteor

Hi all,

 

I built a simple regression model with Linear Regression Tool and when looking at the model statistics I noticed that model statistics from output anchor R produces different numbers compared to numbers from output I. Shouldn't they be exactly same for example in case of Adjusted R Squared.

 

See png for details.

 

Cheers,

 

Juha 

5 REPLIES 5
RolandSchubert
16 - Nebula
16 - Nebula

Hi @JuhaP ,

 

aside from rounding (I: 3 digits, R: 4 digits), the numbers should be same, as it is for F-statistics and coefficient. Haven't had this issue before (or at least not realized). Is it only for this specific model?

 

Best

 

Roland

JuhaP
7 - Meteor

Hi,

 

I'm afraid this is not a first time I notice the difference in these model stats, tested with several models using Linear Regression. I believe stats on the left side of the picture are correct. Tested with, believe or not, in Excel 😉

JuhaP
7 - Meteor

Hi again,

 

Looks like dropping the model constant (alfa) messes the I Output. Strange. 

RolandSchubert
16 - Nebula
16 - Nebula

Exactly that happens. Just reproduced it with one of my models. Seems to be a bug.

JuhaP
7 - Meteor

Hi  

 

 

Labels