| |
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
RE: st: RE: Graphs, pull out random values at peak interval
I don't see where the randomness enters in that
case. Perhaps you just want 25 in the middle
of each panel.
In that case, there are numerous slight variations
on this recipe:
by group, sort : egen maxtime = max(time)
by group : egen mintime = min(time)
gen middletime = floor((mintime + maxtime) / 2)
gen byte select = inrange(time, middletime - 12, middletime + 12)
programthatanalysesyourdataandtellsyoutheirmeaningandgivespublishableresults if select
I assume time measurement on successive days, labelled
with integers.
Nick
[email protected]
Knag Anne-Christine
> As 25 starting from a not randomly selected date until
> another not randomly
> selected date.
On 6/6/07 7:13 PM, "Nick Cox" <[email protected]> wrote:
> > In what sense would these days be randomly selected?
> > As 25 from some larger number? As 25 starting from
> > some randomly selected date? Or what?
> >
> > Nick
> > [email protected]
> >
> > Knag Anne-Christine
> >
> >> I have plotted production for 7 different groups over a
> time period of
> >> around 100 days. There were some days without production,
> >> these days varied
> >> between the groups. Production began and ended at different
> >> days in each of
> >> the groups. I want to compare 25 random days at the middle of
> >> the production
> >> period between and within groups (1factor anova). But how can
> >> I pull out 25
> >> random days in the middle using STATA?
> >>
> >> I have another comparison where I am looking at the max size
> >> of the product
> >> that were measured each day. The max size peaks over a period
> >> of around 7
> >> days. The peaks occurs at different times between the groups.
> >> How can I
> >> compare the sizes within and between the groups during
> just the peak?
*
* 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/