Answered by:
How to get the r squared value from a linear regression model

Hi,
I have created a Linear Regression data mining model and I would like to query the model for some values that are typically linked to a linear regression model:
 R squared
 Adjusted R squared
 Standard Errors
Is it possible to get this values from a DMX query, or do I have to calculate them myself?
Question
Answers

You have to calculate them yourself.
______________________________
Jamie MacLennan
CTO
Predixion Software, Inc. jamiemac@predixionsoftware.com Marked as answer by Jinchun ChenMicrosoft employee, Moderator Monday, March 08, 2010 3:32 AM

While not exposed directly by DMX, these metrics can be easily calculated using a stored procedure.
This article: http://www.sqlserverdatamining.com/ssdm/Default.aspx?tabid=102&Id=8 contains a stored procedure that does exactly this.
bogdan crivat [sql server data mining] / http://www.bogdancrivat.net/dm Marked as answer by Jinchun ChenMicrosoft employee, Moderator Monday, March 08, 2010 3:32 AM
All replies

You have to calculate them yourself.
______________________________
Jamie MacLennan
CTO
Predixion Software, Inc. jamiemac@predixionsoftware.com Marked as answer by Jinchun ChenMicrosoft employee, Moderator Monday, March 08, 2010 3:32 AM

While not exposed directly by DMX, these metrics can be easily calculated using a stored procedure.
This article: http://www.sqlserverdatamining.com/ssdm/Default.aspx?tabid=102&Id=8 contains a stored procedure that does exactly this.
bogdan crivat [sql server data mining] / http://www.bogdancrivat.net/dm Marked as answer by Jinchun ChenMicrosoft employee, Moderator Monday, March 08, 2010 3:32 AM

Hi All,
I have same requirment to calulate
 R squared
 Adjusted R squared
 Standard Errors
but i want to fetch data from cube, can any body tell me, Using Microsoft data mining Linear Regression can we do this calulation.
Thanks & Regaerds,
Manish Kumar Verma