> -----Original Message-----
> From: [email protected] [mailto:owner-
> [email protected]] On Behalf Of Richard Williams
> Sent: Wednesday, August 10, 2005 8:02 PM
> To: [email protected]
> Subject: Re: st: RE: constraints insufficient for unique estimate
>
> At 07:51 PM 8/10/2005 -0500, Scott Merryman wrote:
> >You need to add the -nocons- option if you have full set of dummy
> variables.
> >
> >Scott
>
> Even then, though, I believe he'll have problems if he has full sets of
> dummies for more than 1 categorical variable, won't he?
>
True, good point. One dummy variable for each category would need to be
dropped if therapeutic conditions consist of more than 1 categorical
variable.
This problem is easy enough to replicate:
. sysuse auto
(1978 Automobile Data)
. qui tab fore, gen(Df)
. const define 1 mpg = gear
. cnsreg price mpg gear Df*, con(1)
constraints insufficient for unique estimate
redundant or inconsistent constraints
r(412);
. cnsreg price mpg gear Df* , con(1) nocons
Constrained linear regression Number of obs = 74
<snip>
. qui tab rep, gen(Dr)
. cnsreg price mpg gear Df* Dr*, con(1) nocons
constraints insufficient for unique estimate
redundant or inconsistent constraints
. cnsreg price mpg gear Df* Dr1-Dr4, con(1) nocons
Constrained linear regression Number of obs = 69
<snip>
Scott
*
* 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/
