please disregard my previous question, as i see that the merge, update
command should replace all missing values, including string.
i am, however, still having difficulty implementing this for some reason.
i am trying to merge one dataset with another to add a variable (SIC
codes) to the master set. The SIC variable is not a part of the original
master set. The first time I do merge the two datasets (update command
unnecessary), those companies
that matched in my master set acquire SIC codes while those that do not
have empty values ("" since it is a string variable). In my attempt to
match as many companies as possible, I do a series of match-merges using
increasingly lax criteria to match them. I therefore need subsequent
merges to update the master data with the SIC codes for those entries
where they are missing (I am willing to accept that it will also update
other variables that are missing but I do not wish to use the "replace"
option to force replacements). my log results are showing, however, that
even while hundreds of succesful matches are occuring, no "updating" is.
Indeed, the final dataset only has SIC codes for those succesfully matched
in the first round but not for any others. Might anyone have any
suggestions for what I should be concerned about?
thanks again,
anna
On Sat, 17 Aug 2002, Anna Hyojin Joo wrote:
> might anyone clarify if the merge command with the "update" option will
> replace missing values in the master dataset even when they are string
> variables ("")?
>
> many thanks,
> anna
>
>
>
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
