Re: st: changing beta coefficients

[Lars Kroll <lars@climweb.de>]

Peter Haan schrieb:

> dear stata list,
>
> I have the following question:
>
> Is it possibe to change the estimated vector of beta coefficients by 
> hand.
> I want to predict   with a slightly different beta vector than the  
> estimated,
> The idea is to replace the estimated vector e(b)  by an artificial 
> e(b*) and then to use the command pred,
> however, as far as I know I cannot change e(b).
>
> thanks for your help,
> peter
>
>
> *
> *   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/
Hello Peter,

try this:

sysuse auto,clear

regress  price headroom

matrix __b= e(b)

svmat __b

matrix list __b

* New effect of headroom = b(headroom)/2
* b2 is _cons

gen myprice = __b2 +  (__b1/2)* headroom


Yours,
Lars


--
____________________________
Lars E. Kroll
UNI  030  83857614
RKI  030  45473307
TEL  030  25587610
   0176 23131352
FAX  069  13306720084
WEB www.climweb.de/privat 
*
*   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/