# Re: st: Returning a p-value for simulation

 From Eva Poen To statalist@hsphsun2.harvard.edu Subject Re: st: Returning a p-value for simulation Date Sun, 10 May 2009 09:43:27 +0100

```<>

John,

apologies, I didn't actually write a myprobit before sending the post,
so I guessed about what the variable names after simulation will be
from my previous simulations. It turns out that they would be _b_x1
and _b_x2, not y_b_x1 and y_b_x2 (same for standard deviations).
Therefore, just drop the leading y everywhere in the loop, and
everything should work fine. Here is a working example.

****************
matrix mean = 2,2
matrix sd = 1,4
matrix corr = 1,0.5\0.5,1
capture program drop myprobit
program myprobit
drop _all
set obs 1000
drawnorm x1 x2, mean(mean) sd(sd) corr(corr) cstorage(full)
gen eps = rnormal(0,1)
gen ystar = 5+3*x1-2*x2+eps
gen y = normal(ystar)>0.5

probit y x1 x2
end

simulate _b _se, reps(50) seed(123) : myprobit

foreach v in x1 x2 {
gen t_`v' = _b_`v'/_se_`v'
gen p_`v' = 2*(1-normal(abs(t_`v')))
}

*****************

Hope this helps,
Eva

Eva

2009/5/10 John Antonakis <john.antonakis@unil.ch>:
> Hi Eva:
>
> Working with your solution first--it seems to be simple to use.  So I have
> three RHS variables.  Everything works fine until this point:
>
> . foreach v in x z q {
>  2.   gen t_`v' = y_b_`v'/y_se_`v'
>  3.   gen p_`v' = 2*(1-normal(abs(t_`v'))
>  4. }
> r(111);
>
> Can you tell what's wrong?
>
> Thanks,
> John.
>
> ____________________________________________________
>
> Prof. John Antonakis
> Associate Dean Faculty of Business and Economics
> University of Lausanne
> Internef #618
> CH-1015 Lausanne-Dorigny
> Switzerland
>
> Tel ++41 (0)21 692-3438
> Fax ++41 (0)21 692-3305
>
> Faculty page:
> http://www.hec.unil.ch/people/jantonakis&cl=en
>
> Personal page:
> http://www.hec.unil.ch/jantonakis
> ____________________________________________________
>
>
>
> On 10.05.2009 09:53, Eva Poen wrote:
>>
>> <>
>>
>> If John wishes to return coefficients and associated statistics after
>> estimation for a simulation he might want employ an eclass program.
>> This way, he can easily access coefficients and standard errors, and
>> calculate t-statistics and p-values afterwards. Here is an example:
>>
>>
>> *******************
>> program myprobit, eclass
>>   ...
>>   probit y x1 x2
>> end
>>
>> simulate _b _se , reps(1000) seed(123) saving(myprobit, replace
>> every(50)) : myprobit
>>
>> use myprobit
>>
>> foreach v in x1 x2 {
>>  gen t_`v' = y_b_`v'/y_se_`v'
>>  gen p_`v' = 2*(1-normal(abs(t_`v'))
>> }
>> **********************
>>
>> Hope this helps,
>> Eva
>>
>>
>> 2009/5/9 John Antonakis <john.antonakis@unil.ch>:
>>
>>>
>>> I am running an rclass program that I wish to simulate.
>>>
>>> After estimating a probit, I want to save the coefficients, along with
>>> the
>>> t-stats and the p-values for later simulation.  I figured out the former
>>> but
>>> not the latter two. What I have so far is:
>>>
>>> probit y x1 x2
>>> return scalar b1 =_b[x1]
>>> return scalar b2 =_b[x2]
>>>
>>> Thanks,
>>> John.
>>>
>>> --
>>> ____________________________________________________
>>>
>>> Prof. John Antonakis
>>> Associate Dean Faculty of Business and Economics
>>> University of Lausanne
>>> Internef #618
>>> CH-1015 Lausanne-Dorigny
>>> Switzerland
>>>
>>> Tel ++41 (0)21 692-3438
>>> Fax ++41 (0)21 692-3305
>>>
>>> Faculty page:
>>> http://www.hec.unil.ch/people/jantonakis&cl=en
>>>
>>> Personal page:
>>> http://www.hec.unil.ch/jantonakis
>>> ____________________________________________________
>>>
>>> *
>>> *   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/
>>
>
> *
> *   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/
```