Re: st: regress and then extrapolate with new dependent vars but same coefficients

[Richard Williams <Richard.A.Williams.5@ND.edu>]

At 09:56 AM 7/19/2006, Olga Gorbachev Melloni wrote:
> Hello,
>
> I need to run an exercise in which after running a regression and 
> predicting the y_hat, I need to replace one of the dependent 
> variables with zeros and then use the saved coefficients to predict 
> a hypothetical value for y_hat(given x1==0), in order to compare the 
> predicted values from these two exercises.
>
> Does any one know how to do this quickly, without saving all the 
> coefficients and manually predicting for the second step? I am 
> running a regression with 174 dependent variables (mostly dummies) 
> and it would take some time to predict manually.
>
> I've checked on the faq for this and was unable to find anything, i 
> might have overlooked the answer and apologize in advance if it has 
> been asked and answered already and would appreciate any references if so.
One way is to use the adjust command.

reg y x1 x2 x3
predict yhat1
adjust x1=0, gen(yhat2)

Another way:

clonevar newx1 = x1
reg y newx1 x2 x3
predict yhat1
replace newx1 = 0
predict yhat2


-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
FAX:    (574)288-4373
HOME:   (574)289-5227
EMAIL:  Richard.A.Williams.5@ND.Edu
WWW (personal):    http://www.nd.edu/~rwilliam
WWW (department):    http://www.nd.edu/~soc 
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/