Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Can I change the intercept in a linear regression model


From   "JVerkuilen (Gmail)" <[email protected]>
To   [email protected]
Subject   Re: st: Can I change the intercept in a linear regression model
Date   Sat, 20 Apr 2013 10:11:46 -0500

On Sat, Apr 20, 2013 at 6:57 AM, Daniel Herbert Opi
<[email protected]> wrote:
>
> I would like to run this model with the intercept instead of running
> through zero it runs through a specified value i.e the mean of my
> continuous variable.
>
> Is there a way of doing this in Stata?

Stata itself has nothing to do with it per se.

The intercept is the predicted value of the dependent variable when
all X variables are 0. So if you move the 0 for the X variables you
move the intercept (but not the slopes). Thus if you subtract out
whatever value you want from the relevant X variable you make the
intercept interpretable, assuming that 0 for the X variable is not
itself interpretable. I often recommend in class that all continuous
predictors be centered, either on the sample mean or some other
relevant number, such as a population mean or scale norm if one is
available. Example: If you were using an IQ test score as a predictor,
it is typically the case that IQ tests have a mean of 100 and standard
deviation of 15 by convention. Thus is might make sense to subtract
100 from all scores, not the sample mean. There are sometimes reasons
not to, but it's usually pretty good practice.

Basically just make the new variables you want using generate and egen.
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index