# st: Re: quick question on lincom

 From "Scott Merryman" To Subject st: Re: quick question on lincom Date Fri, 20 Dec 2002 16:34:06 -0600

```One can always check the results of lincom by hand...

. clear

. use "C:\Stata\auto.dta", clear
(1978 Automobile Data)

. qui reg price mpg gear

. lincom mpg + gear

( 1)  mpg + gear_ratio = 0.0

------------------------------------------------------------------------------
price |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
(1) |  -485.9787   819.8677    -0.59   0.555    -2120.748    1148.791
------------------------------------------------------------------------------

. vce

|      mpg gear_r~o    _cons
-------------+---------------------------
mpg |  4599.44
gear_ratio | -35934.6   739453
_cons |  10382.5 -1.5e+06  4.3e+06

. disp sqrt(4599.44 + 739453 +  2*(-35934.6))
819.86782

Scott

----- Original Message -----
From: "Kihong Eom" <keom0@uky.edu>
To: <statalist@hsphsun2.harvard.edu>
Sent: Friday, December 20, 2002 3:05 PM
Subject: st: quick question on lincom

> Dear Statalist:
>
> Is "lincom" taken into account of covariance between the estimated
> parameters?
> Let's suppose the linear combination of b1 + b2.  The combined
> coefficient is b1 + b2.
>
> The question is the variance of (b1 + b2) is V(b1, b2) or V(b1+b2)?
> This is different, because V(b1, b2) = V(b1) + V(b2) + 2C(b1, b2), yet
> V(b1+b2) = V(b1+b2).  I think the variance for lincom is the latter.  If
> so, do you know any written program for the former?
>
>
> Kihong
>

*
*   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/
```