Re: st: suest after ivregress or ivreg2

[Austin Nichols <austinnichols@gmail.com>]

Try this.

est clear
sysuse auto, clear
ren weight z
ren headroom x
ivregress 2sls price (x=z)
estimates store one
ren x headroom
ren turn x
ivregress 2sls price (x=z) mpg
estimates store two
ren x turn
preserve
g i=_n
expand 2
bys i: g n1=_n-1
g x1=headroom*!(n1==0)
g z1=z*!(n1==0)
g x2=turn*!(n1==1)
g z2=z*!(n1==1)
replace mpg=0 if n1==1
g n2=1-n1
ivregress 2sls price (x1 x2=z1 z2) mpg n1 n2, cl(i) nocons
est sto stacked
test x1=x2
testnl _b[x1]*_b[x2]=_b[mpg]
restore
esttab one two stacked, nogaps mti


On Wed, Apr 7, 2010 at 3:54 PM, richard boylan <richardtb25@gmail.com> wrote:
> Ok. This is what I am trying to do.
>
> sysuse auto.dta
> ivregress 2sls price (headroom = weight) mpg
> estimates store one
> ivregress 2sls price (turn = weight)
> estimates store two
> suest one two
> testnl [one]headroom*[two]turn = [one]mpg
>
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/