Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

RE: st: RE: how to generate a variable using the estimation results


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: st: RE: how to generate a variable using the estimation results
Date   Fri, 3 Nov 2006 16:43:31 -0000

This is standard stuff covered in the help and
the manuals or learnable by simple experiments. 

But in short 

_b[] 

is Stata standard notation for a parameter (coefficient) in 
a model. Usually a varname belongs in the [], with 
the important exception _cons. But with non-linear 
models there is typically not a one-to-one correspondence 
between parameters and variables -- else they would be 
linear -- and so the notation /parametername is the 
alternative adopted. Or so I surmise. 

Try 

. sysuse auto
. gen gpm = 1/mpg
. nl (gpm = {a}*weight) 

or any other example. (I know it's linear). 


Nick 
n.j.cox@durham.ac.uk 

Kyle Hood

> what does the notation _b[/rho] do?  Why _b[]?  why /?

Oleksandr Shepotylo wrote:
> 
> > Thank you,
> >
> > I need this rho to transform my exogenous variables X(NxK) 
> by mulitplying 
> > them by the following matrix:
> >
> > X*=rho*W*X, where W is (NxN)
> >
> > so I planned to calcualte rho*X and use mata to calculate W*(rho*X).
> >
> > But you are right, I can save a scalar rho and use it in 
> mata expressions 
> > when I need it.

Nick Cox wrote:

> >> This sounds like a bad idea. The parameter can be saved as 
> a scalar or a 
> >> local macro, but using a variable for
> >> the same number again and again is usually unnecessary. 
> >> Still, there might be a good idea, or you might insist, so 
> if so try 
> >> gen rho = _b[/rho] 
> >> but consider also
> >> 
> >> scalar rho = _b[/rho] 

Oleksandr Shepotylo

> >>> After non linear estimaition which estimates parameters 
> rho and sigma of 
> >>> the following form
> >>> 
> >>> nl (z=h1*{rho}+h2*{rho}^2+h3*{sigma2}), in(rho -0.5 
> sigma2 1) iterate(200)
> >>> 
> >>> I want to generate a variable rho that takes the value of 
> the estimated 
> >>> parameter {rho}
> >>> 
> >>> gen rho=???

*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index