[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: RE: vectored data?
Your data sound like three separate variables:
after which you can go
gen volume = length * breadth * height * _pi / 6
if you wish.
The fact that your variables hang together in a
strong sense is something you know, but there is neither need
nor possibility to tell Stata. All variables in Stata
are on the same footing, with no scope
for indicating links between them.
Stata has object-oriented programming. I guess you
don't need it at all for what you want.
A mouse, similarly, corresponds to one or more
observations (cases, records) in your dataset.
> I've searched around the Stata resources for some time now but
> haven't been able to find a solution for my problem (which I hope
> isn't a FAQ, or perhaps even RTFM).
> I'm doing statistics on an animal experiment that I've just
> Tumor bearing mice were submitted to different treatments ordered
> into treatment groups. Respons to treatment was quantified by
> measuring tumor size daily, however not in weekends. Tumor
> subqutaneously in the foot) size was measured as length, breadth and
> height, and the volume calculated as length*hight*breadth*\pi/6. I
> haven't been able to find a satisfactory way to enter this data into
> a Stata dta file. I'd like to be able to enter a data point as a 3-d
> vector, but I don't know how, since this isn't a data type. My
> current solution is to enter each measurement as a text string:
> "ll,hh,bb" and then extract each parameter when needed. Is there a
> 'smarter' way?
> Also, is it possible to define a 'class' or 'object' as a data type
> (couldn't find anything appropriate in the 'Programming' manual)
> called Mouse containing all this kind of information (mouse number,
> group number, treatment date, daily sizes, etc.)?
* For searches and help try: