Re: programming syntax problem [was: Re: st: From: "Christopher L. Aberson" ...]

[Nick Cox <njcoxstata@gmail.com>]

Now a fuller answer by Maarten Buis at

http://stackoverflow.com/questions/17156909/user-program-not-recognized-in-simulation-command-in-stata
Nick
njcoxstata@gmail.com


On 18 June 2013 08:14, Nick Cox <njcoxstata@gmail.com> wrote:
> If you look again at the help for -simulate- the main example is a
> program with options. The -simulate- call shows that program being
> called with arguments for its options.
>
> No option calls are given in your -simulate- call. This seems
> especially important as you have required options in -powersim3-. How
> is -powersim3- supposed to operate then?
>
> For my part, I don't understand why this basic point is not clear to
> you. If this doesn't help, or I in turn am misunderstanding, sorry,
> but I end there.
>
> Nick
> njcoxstata@gmail.com
>
>
> On 18 June 2013 05:11, Christopher L. Aberson
> <Christopher.Aberson@humboldt.edu> wrote:
>> Respectfully -- I don't understand what needs to be supplied. That's
>> why I've been posting here. No matter how many times you say "look at
>> this or that" I'm not going to be able to figure this out on my own.
>> That is why I am seeking help.
>>
>>
>>
>>>Look again at your line
>>>
>>>simulate _b _se, reps(100): powersim3
>>>
>>>-powersim3- has compulsory options but you don't supply them.
>>>
>>>Nick
>>>njcoxstata@gmail.com
>>
>>
>> On 18 June 2013 00:39, Christopher L. Aberson
>> <Christopher.Aberson@humboldt.edu> wrote:
>>> Not sure I follow you. If I run this without using the single command line
>>> approach (i.e., just plug in numbers I want here) the Simulate command is
>>> getting fed.
>>>
>>> clear all
>>> program define powersim3
>>> drop _all
>>> matrix r = (1.0,.30,.30\.30,1.0,.30\.30,.30,1.0)
>>> matrix m = (0,0,0)
>>> matrix sd = (1,1,1)
>>> drawnorm y x1 x2, n(282) corr(r) means(m) sds(sd)
>>>
>>> regress y x1 x2
>>> end
>>>
>>> simulate _b _se, reps(100): powersim3
>>>
>>> Yet when I try to execute this using a command line as shown in the previous
>>> posts, simulate isn't getting fed.
>>>
>>> Can you clarify how I can tweak the original code to make this run?
>>>
>>>
>>> <You are using -simulate- to call your -powersim3-, but not feeding it
>>> anything.
>>> <Nick
>>> <njcoxstata@gmail.com
>>>
>>>
>>> On 17 June 2013 21:56, Christopher L. Aberson
>>> <Christopher.Aberson@humboldt.edu> wrote:
>>>> Thank you for the feedback. The approaches suggested made the code
>>>> more elegant but did not solve the core issue.
>>>>
>>>> To recap, I wand to run the code noted below (encased in *****) using
>>>> this line powersim3, ry1(.30) ry2(.30) r12 (.30) n (282)
>>>> that defines the required characteristics.
>>>>
>>>> It appears the issue is that the simulate command isn't recognizing
>>>> the powersim3 command above. I expect there is a simple solution here
>>>> -- but I'm not proficient enough in Stata to know what the simple
>>>> answer is. Any additional suggestions would be very helpful.
>>>>
>>>> ******
>>>>  program define powersim3, rclass
>>>>  syntax, ry1(real)ry2(real)r12(real)n(
>>>> real) [my(real 0)m1(real
>>>> 0)m2(real 0)m3(real 0)sy(real 1) ///
>>>>     s1(real 1)s2(real 1)s3(real 1)alpha(real .05)help]
>>>> matrix m = (`my', `m1', `m2')
>>>> matrix sd = (`sy', `s1', `s2')
>>>> matrix r = (1, `ry1', `ry2'\ `ry1' , 1 , `r12'\ `ry2' , `r12', 1)
>>>> drawnorm y x1 x2, n(`n') corr(r) means(m) sds(sd)
>>>> regress y x1 x2
>>>> end
>>>>
>>>> simulate _b _se, reps(10000): powersim3
>>>> *****
>>>>
>>>>
>>>>> Please use meaningful titles for your postings.
>>>>>
>>>>> Try editing out the spaces between option names and option details in
>>>>> the cases of the options
>>>>>
>>>>> ry1 (real) ry2 (real) r12 (real) sy (real 1)   s1 (real 1) s2 (real 1)
>>>>> s3 (real 1)
>>>>>
>>>>> Incidentally, your program does nothing with the options
>>>>>
>>>>> alpha(real .05) help
>>>>>
>>>>> although you may be planning to add details later.
>>>>>
>>>>> Nick
>>>>> njcoxstata@gmail.com
>>>>>
>>>>>
>>>>> On 14 June 2013 23:18,  <owner-statalist@hsphsun2.harvard.edu> wrote:
>>>>>> I've written a tiny program for simulation of multiple regression
>>>>>> results.  I can run this on a line by line basis (using actual values
>>>>>> instead of the syntax command), but when I put it in an ado, it
>>>>>> doesn't work.
>>>>>>
>>>>>> The basic idea is use drawnorm to generate a dataset, run regression
>>>>>> on it, then do it over and over using the simulate option.
>>>>>>
>>>>>>
>>>>>> program define powersim3, rclass
>>>>>> syntax, ry1 (real) ry2 (real) r12 (real) [my(real 0) m1(real 0)
>>>>>> m2(real 0) m3(real 0)] [sy (real 1) ///
>>>>>>     s1 (real 1) s2 (real 1) s3 (real 1)] n(real) [alpha(real .05) help ]
>>>>>> matrix m = (`my', `m1', `m2')
>>>>>> matrix sd = (`sy', `s1', `s2')
>>>>>> matrix r = (1, `ry1', `ry2'\ `ry1' , 1 , `r12'\ `ry2' , `r12', 1)
>>>>>> drawnorm y x1 x2, n(`n') corr(r) means(m) sds(sd)
>>>>>> regress y x1 x2
>>>>>> end
>>>>>>
>>>>>> simulate _b _se, reps(10000): powersim3
>>>>>>
>>>>>> .powersim3, ry1(.30) ry2(.30) r12 (.30) n (282)
>>>>>>
>>>>>> This yields an error saying "ry1" required. If I take out the simulate
>>>>>> command I get a single regression but not the repeated
>>>>>> sampling/simulation I want.>
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/