Link to home
Start Free TrialLog in
Avatar of RogueStat
RogueStat

asked on

LINEST Limit

Hey guys,
I have a LINEST question.  Is there a limit to the number of known x's I can put into LINEST?  

I've hit #REF! errors when using 17 plus an intercept (or more).  It worked for 15 plus an intercept (I don't have a case with 16 plus int.)

I remember something about a limit in Excel 2003 but I wasn't sure for 2007.  Anyone know for sure?
ASKER CERTIFIED SOLUTION
Avatar of Patrick Matthews
Patrick Matthews
Flag of United States of America image

Link to home
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Start Free Trial
BTW, I *strongly* discourage using Excel for regression analysis for anything that is remotely important:

1) LINEST (and yes, the Analysis ToolPak uses LINEST for its regression calculations) will occasionally return incorrect results, such as the mathematically-impossible negative r-squared

2) Excel does not test for true independence amongst the supposed independent variables

For any statistical analysis that really matters, I use the (pricey but excellent) Minitab.
Avatar of RogueStat
RogueStat

ASKER

Thanks Matthew, I appreciate it.

I'm going to use something more robust like Minitab for the "real" analysis.  It's going to have a lot more data, so Excel would probably bog down even if it didn't produce some head-scratching errors.

Right now I'm just quickly demonstrating a concept.  Isn't all of the regression based off of Solver in Excel or did I just decide they were the same thing?
You confirmed my suspicion, thank you!
Solver is an independent product licensed by Microsoft for inclusion in Excel. A newer version has been licensed for Excel 2010. Neither the old nor the new version is used for regression analysis.

Brad