--- Janet Oliver <[email protected]> wrote:
> I have a dataset that consists of a name (of various
> lengths), a treatment
> and a time period (0..5) with not all patients being
> seen at the six times,
> and various clinical results. I wanted to create a
> unique numerical
> identifier based on name|treatment|time. I initally
> concatenated the
> variables "egen tid=concat(patient treat time)" and
> the tried to destring
> tid but this failed with the error message "tid
> contains non-numeric
> characters; no generate". I have examined the
> variables in the editor and
> cannot see any non-numeric characters.
> My questions are 1) Is there any way of determining
> non-numeric characters
> or 2) Is there an efficient method of combing
> name|treatment|time and
> generating a unique ID (preferably starting at 1 and
> increasing
> sequentially) - or should I just type in the IDs by
> hands.
> Many thanks.
> Janet
Janet, if you haven't done it already you can use
"describe" to confirm that the variables are numeric.
Although they may look like numbers the may actually
be stored as character, depending on how the data set
was created. To convert a character to a numeric, you
can use the "real" function, e.g., gen
num=real(string)
To help you identify unique observations, you may be
able to use explicit subscripting, e.g. "_n", see the
USer's Manual, or help for subscripting.
__________________________________
Do you Yahoo!?
Find out what made the Top Yahoo! Searches of 2003
http://search.yahoo.com/top2003
*
* 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/
