Re: st: Metareg syntax error for wsse option

["Jon Nelson" <jpn@psu.edu>]

Sorry, it is not a typo on my part if that what's you are implying. But i think
something in the original metareg command got corrupted by changes in Stata. I
got out my original disk to Stata 11, and tried to use it to "repair" the
program files. Somehow --don't ask me how -- i ended up with two versions of
Stata 11 -- one labeled Stata-64_old (dated 09/01/11) and a second labeled
Stata-64 (dated 02/16/12). The old Stata works with metareg and the new Stata
does not. 

Signing off,
Jon

On Tue, Nov 20, 2012 03:13 PM Nick Cox <njcoxstata@gmail.com> wrote:
>
>That is what the syntax diagram says, but the code and examples contradict it.
>
>Hence my assertion that it's a typo.
>
>Nick
>
>On Tue, Nov 20, 2012 at 7:39 PM, Rebecca Pope <rebecca.a.pope@gmail.com> wrote:
>> I think your command should be: metareg vsl09 loginc09 wsse(sd09), mm.
>> Note, wsse() comes before the comma; this is why you are getting an
>> error message about it not being an option.
>>
>> Here is the syntax from the help file for -metareg-:
>>  metareg depvar [indepvars] [if] [in] wsse(varname) [, eform graph
>>                randomsize noconstant mm reml eb knapphartung z tau2test
>>                level(#) permute(# [, univariable detail joint(varlist1 [|
>>                varlist2 ...])]) log maximize_options]
>>
>>
>> Rebecca
>>
>>
>>
>>          __o                __o
>>       _`\ <,_            _`\ <,_
>>      (_)/   (_)          (_)/   (_)
>> ========================>
>>
>> On Mon, Nov 19, 2012 at 1:24 PM, Nick Cox <njcoxstata@gmail.com> wrote:
>>>
>>> Jon Nelson replied to me privately. For several reasons why you are
>>> asked not to do that, see
>>>
>>> http://www.stata.com/support/faqs/resources/statalist-faq/#private
>>>
>>> Most obviously, no one else can see your post. Also, replying to your
>>> email does not constitute a promise of support until the problem is
>>> solved!
>>>
>>> Programs don't suddenly stop working. There is some other explanation,
>>> but I am sorry that I don't know what it is.
>>>
>>> The program can be downloaded from both SSC and SJ. The two versions
>>> look identical. Nevertheless you should, as asked originally, explain
>>> _where_ you got your copy from. (Jon should confirm that he is _not_
>>> using the ancient version from STB-42.)
>>>
>>> There is a typo in both help files in that -wsse()- is listed in the
>>> syntax diagram as something that should be typed before the comma.
>>> This is not biting Jon as he  appears to be using the option
>>> correctly. However, I have not used -metareg- in any of its
>>> incarnations and can't advise further.
>>>
>>> If a better answer is not forthcoming, I would contact the program
>>> author directly.
>>>
>>> Nick
>>>
>>> ---------- Forwarded message ----------
>>> From: Jon Nelson <jpn@psu.edu>
>>> Date: Mon, Nov 19, 2012 at 6:38 PM
>>> Subject: Re: st: Metareg syntax error for wsse option
>>> To: Nick Cox <njcoxstata@gmail.com>
>>>
>>>
>>> Thanks, but removing spaces did not help. I still get the same error
message.
>>> Here is the exact version of a command:  metareg vsl09 loginc09, wsse(sd09)
mm
>>>
>>> which worked 6 months ago and now returns the error message: "option wsse()
not
>>> allowed" r(198);
>>>
>>> Jon
>>>
>>> On Mon, Nov 19, 2012 12:53 PM Nick Cox <njcoxstata@gmail.com> wrote:
>>> >
>>> >-metareg- is from SJ and SSC. -metareg- in Stata terms is a command,
>>> >not a macro.
>>> >
>>> >Exactly what you typed can only be guessed if you report that it was
>>> >"something like" your example.
>>> >
>>> >It may be that you are putting a space in "wsse ()". If so, remove the
space.
>>> >
>>> >Nick
>>> >
>>> >On Mon, Nov 19, 2012 at 4:54 PM, Jon Nelson <jpn@psu.edu> wrote:
>>> >
>>> >> I'm using the macro metareg to estimate random-effects regressions, using
>>> data
>>> >> that i used successfully six months ago. Now i'm getting an
>>> >> error message "option wsse () not allowed" followed by r(198). This
seems to
>>> >> indicate a syntax error, but i can't see what i'm doing wrong or
>>> >> determine other sources of error.
>>> >>
>>> >> My command line would be something like "metareg vsl09 logincome,
wsse(sd09)
>>> >> mm" , where sd09 is the variable for standard errors of the effect size
>>> vsl09.
>>> >> I tried using other data sets, and i get the same error. I tried
reloading
>>> the
>>> >> macro. I checked the data for zeros or negative values. And wsse () is
>>> >> a required option for metareg.
>*
>*   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/
>
>


Jon Nelson



*
*   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/