Quantcast

suest after quantile regression

classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

suest after quantile regression

xueliansharon
Dear all,

I want to test the equality of coefficients from two quantile regression models, but -suest- doesn't support after the -qreg-, I wonder whether "stacking" method of Austin Nichols http://www.stata.com/statalist/archive/2009-11/msg01485.html can be applied to this case.

That is:

the models I want to test are like:

est clear
sysuse nlsw88, clear
qreg wage hours age if south==1, quantile(10)
est sto south
reg wage hours age if south==0, quantile(10)
est sto nonsouth
suest south nonsouth

however, -suest- doesn't work after -qreg-, so I want to use the "stacking" method as:

est clear
sysuse nlsw88, clear
gen nonsouth=0
replace nonsouth=1 if south==0
expand 2
bys idcode: g n=_n
keep if (n==1&south)|(n==2&nonsouth)
gen hours1=hours*!(n==1&south)
gen hours2=hours*!(n==2&nonsouth)
gen age1=age*!(n==1&south)
gen age2=age*!(n==2&nonsouth)
qreg wage hours1 hours2 age1 age2 n, quantile(10)
test (_b[hours1]=_b[hours2]) (_b[age1]=_b[age2])

Is this procedure correct? I am not sure whether there's any problem to use "stacking" method after quantile regression.

Any help will be greatly apprecitated.

Thanks,
Sharon




Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: suest after quantile regression

Jorge Eduardo Pérez Pérez
Basically you're trying to test whether the coefficients are the same
for two groups, so I think this would work:

sysuse nlsw88, clear
xi: qreg wage i.south*hours i.south*age, quantile(10)
test  _IsouXhours_1=0, notest
test  _IsouXage_1=0, accum


_______________________
Jorge Eduardo Pérez Pérez




On Fri, Aug 27, 2010 at 2:34 PM, xueliansharon <[hidden email]> wrote:

> Dear all,
>
> I want to test the equality of coefficients from two quantile regression
> models, but -suest- doesn't support after the -qreg-, I wonder whether
> "stacking" method of Austin Nichols
> http://www.stata.com/statalist/archive/2009-11/msg01485.html can be applied
> to this case.
>
> That is:
>
> the models I want to test are like:
>
> est clear
> sysuse nlsw88, clear
> qreg wage hours age if south==1, quantile(10)
> est sto south
> reg wage hours age if south==0, quantile(10)
> est sto nonsouth
> suest south nonsouth
>
> however, -suest- doesn't work after -qreg-, so I want to use the "stacking"
> method as:
>
> est clear
> sysuse nlsw88, clear
> gen nonsouth=0
> replace nonsouth=1 if south==0
> expand 2
> bys idcode: g n=_n
> keep if (n==1&south)|(n==2&nonsouth)
> gen hours1=hours*!(n==1&south)
> gen hours2=hours*!(n==2&nonsouth)
> gen age1=age*!(n==1&south)
> gen age2=age*!(n==2&nonsouth)
> qreg wage hours1 hours2 age1 age2 n, quantile(10)
> test (_b[hours1]=_b[hours2]) (_b[age1]=_b[age2])
>
> Is this procedure correct? I am not sure whether there's any problem to use
> "stacking" method after quantile regression.
>
> Any help will be greatly apprecitated.
>
> Thanks,
> Sharon
>
>
>
>
>
> --
> View this message in context: http://statalist.1588530.n2.nabble.com/suest-after-quantile-regression-tp5470620p5470620.html
> Sent from the Statalist mailing list archive at Nabble.com.
> *
> *   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/
>

*
*   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/
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: suest after quantile regression

xueliansharon
Hi Jorge,

Thanks for your suggestion. It's helpful.

Best,
Sharon
Loading...