Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Unable to clear "invalid syntax r(197);" error in user-written .ado file


From   Nick Cox <[email protected]>
To   [email protected]
Subject   Re: st: Unable to clear "invalid syntax r(197);" error in user-written .ado file
Date   Fri, 5 Oct 2012 17:39:29 +0100

You're welcome. As a further footnote, note that your previous program
made -byvar()- an optional option, but would in fact fail if it were
not specified.

That is, with no -byvar()- specified,

sort `byvar'

would just be interpreted as

sort

which is not acceptable to Stata, and the program would fail.

Of course, if the program were just to be used by you and you would
always specify -byvar()- then you would not be bitten in practice.

Nick

That's no longer true of the version below, although I can't claim I
was thinking about this detail when I wrote it.

On Fri, Oct 5, 2012 at 2:53 PM, Richard Herron
<[email protected]> wrote:
> Works like a charm! Thanks for the education, Nick and Joe.
>
> On Fri, Oct 5, 2012 at 5:23 AM, Nick Cox <[email protected]> wrote:
>> Your program accepts -if- and -in-, but does nothing about them.
>> -marksample- is the way to do this.
>>
>> Setting aside the string default issue answered by Joseph, a revised
>> program might be something like
>>
>> program atrim
>>      version 11.2
>>      syntax varlist(numeric) [if] [in] ///
>>          [, Byvar(string) Tail(real 0.5) Suffix(string)]
>>
>>     quietly {
>>
>>      marksample touse, novarlist
>>      count if `touse'
>>      if r(N) == 0 error 2000
>>
>>     tempvar thisuse pct
>>     gen byte `thisuse' = .
>>
>>      foreach x of local varlist {
>>          replace `thisuse'  = `touse' & !missing(`x')
>>          bysort `thisuse' `byvar': egen `pct' = rank(`x')
>>          bysort `thisuse' `byvar' : replace `pct' = 100 * (`pct' - 0.5) / _N
>>          generate `x'_`suffix' = `x' if `thisuse' & inrange(`pct',
>> `tail', 100 - `tail')
>>          drop `pct'
>>      }
>>   }
>>
>>    end
>>
>> On Fri, Oct 5, 2012 at 3:24 AM, Richard Herron
>> <[email protected]> wrote:
>>> I wrote an .ado file to trim data in the left and right tails, but I
>>> can't get past an "invalid syntax r(197)" error. My -syntax- line is
>>> as follows.
>>>
>>> * begin syntax line
>>>     syntax varlist(numeric) [if] [in] ///
>>>         [, Byvar(string fyear) Tail(real 0.5) Suffix(string tr)]
>>> * end syntax line
>>>
>>> I would like to use a -varlist- so that I can loop over the supplied
>>> variables, but I can't clear the error using -varname- or by dropping
>>> the -if- and -in- statements. I also tried making the options required
>>> without defaults, but that didn't work either.
>>>
>>> What is my misunderstanding of syntax? Thank you!
>>>
>>> Below is the complete .ado file. I use Stata SE 11.2 (64 bit) on Windows 7.
>>>
>>> * begin .ado file
>>> *! 0.1 Richard Herron 10/3/2012
>>> program atrim
>>>     version 11.2
>>>     syntax varlist(numeric) [if] [in] ///
>>>         [, Byvar(string fyear) Tail(real 0.5) Suffix(string tr)]
>>>
>>>     * rather than -bysort- twice below
>>>     sort `byvar'
>>>
>>>     * not necessary to tokenize; loop varlist
>>>     foreach x of local varlist {
>>>
>>>         * determine percentiles
>>>         by `byvar': egen num = count(`x') if !missing(`x')
>>>         by `byvar': egen ran = rank(`x') if !missing(`x')
>>>         generate pct = 100 * (ran - 0.5) / num
>>>
>>>         * replace tails with -sysmiss-
>>>         generate test = if inrange(pct, `tail', `=100 - `tail'')
>>>         generate `x'_`suffix' = cond(test, `x', .)
>>>
>>>         * drop "temporary" variables
>>>         drop num ran pct test
>>>
>>>     }
>>>
>>>     end
>>> * end .ado file
>> *
>> *   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/
> *
> *   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/
*
*   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/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index