Quinn-Curtis Forums
Quinn-Curtis Forums
Home | Profile | Register | Active Topics | Members | Search | FAQ
Save Password
Forgot your Password?

 All Forums
 Tools for Microsoft .Net & .Net Compact Framework
 QCMatPack Matrix and Analysis Tools (C# and VB)
 binary variable using MulReg
 New Topic  Reply to Topic
 Printer Friendly
Author  Topic Next Topic  


1 Posts

Posted - 09 Sep 2009 :  15:49:36  Show Profile  Reply with Quote
How would you represent a variable with values of 1 or 0 in a multiple regression using the StepwiseMulReg class? If I set the value to 0, then the entire set of observations is thrown out. Along the same lines, if a value really is zero (say a percentage) then how would you be able to include it in the regression.

Is there something I'm missing?



1569 Posts

Posted - 09 Sep 2009 :  16:02:02  Show Profile  Reply with Quote
We aren't exactly sure what you mean.

Are you saying that you have an independent variable, that has a value of either 0, or 1 (no fractions)? And when you run a stepwise regression the variable is always deemed insignificant and left out of the final equation?

Is the binary variable significant? How do you know? If you run a standard multiple regression (NOT stepwise) on the data, including all variables, what is the F-STAT value for the binary independent variable.
Go to Top of Page
   Topic Next Topic  
 New Topic  Reply to Topic
 Printer Friendly
Jump To:
Quinn-Curtis Forums © 2000-2018 Quinn-Curtis, Inc. Go To Top Of Page
Powered By: Snitz Forums 2000 Version 3.4.07