Re: st: reclink type mismatch error

[Richard Herron <richard.c.herron@gmail.com>]

Michael guided me to explicitly strip out special characters. This
cleared my error and completed my match.

To ensure that I only had letters and spaces I used -sieve()- from the
-egenmore- package from SSC with the option -keep(alphabetic space)-.
The -keep(alphabetic space)- option removes all characters that aren't
A-Z, a-z, or spaces, which is sufficient for my matching on names.

On Wed, Oct 24, 2012 at 11:10 PM, Richard Herron
<richard.c.herron@gmail.com> wrote:
> With the updated -reclink- (Thanks, Michael!) I get a different error.
>
> Going through 166316 observation to assess fuzzy matches, each .=5% com
>> plete
> ...........too few quotes
> r(132);
>
> I am using the same master and using data. I am comfortable discarding
> observations if doing so generates some matches. Is there a
> -subinstr()- or filter that I can apply to get past this error?
> Thanks!
>
> On Wed, Oct 24, 2012 at 4:10 PM, Michael Blasnik
> <michael.blasnik@gmail.com> wrote:
>> Sorry I haven't gotten around to sending the updated reclink to SSC.
>> I will email a copy directly to the poster.
>>
>> Michael Blasnik
>>
>>
>> On Tue, Oct 23, 2012 at 2:38 PM, Richard Herron
>> <richard.c.herron@gmail.com> wrote:
>>> I get a "type mismatch" error using the current version of SSC's
>>> -reclink-. This is the same error as a post from two weeks ago
>>> (http://www.stata.com/statalist/archive/2012-10/msg00290.html -- new
>>> code from the author resolved this error). The call and error are:
>>>
>>> * begin console output
>>> . reclink initials full_name first_name last_name  ///
>>>>     using $DATADIR/finance_faculty, ///
>>>>     idmaster(index_rm) idusing(index_ff) gen(score) ///
>>>>     orblock(initials)
>>>
>>> 48 perfect matches found
>>>
>>> Going through 166316 observation to assess fuzzy matches, each .=5% complete
>>> .............type mismatch
>>> r(109);
>>> * end console output
>>>
>>> For what it's worth, I converted the using file from unicode to ascii
>>> using Python (it is scraped from a website) before generating the .dta
>>> file in Stata. I converted the master file from a SAS data file to a
>>> Stata data file using SAS (it comes from Wharton's WRDS database), but
>>> I am not sure of its encoding.
>>>
>>> Is there a way to guarantee the master data file is ascii, also? Or is
>>> there a way to troubleshoot this error or convert the offending
>>> characters without knowing them a priori?
>>>
>>> Please let me know if any other information would be helpful.
>>>
>>> Thank you!
>>> *
>>> *   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/