--- On Thu, 10/9/09, Muhammad Riaz wrote privately:
> I am stuck at a point, I wandered if I can ask for
> your help with stata syntax.
> I have got a total score for patients over 4 periods
> of time. the graphical display shows that there
> are some patients with scores that
> increase over time, others with scores that decrease, and
> others with scores that remain approximately the
> same.
>
> I want to create three
> groups:
> 1=subjects with total score
> decreasing over time, 2=subjects with total score decreasing
> over time, 3=subjects with total score remains
> approximately the same over time.
First off, don't sent these type of questions privately, but
to the Statalist, as is discussed in the Statalist FAQ:
http://www.stata.com/support/faqs/res/statalist.html#private
The conceptually tricky point in your question is to define
"approximately the same", in the example below I defined that
as a non-significant linear trend at the 10% level.
*--------------- begin example --------------
clear
input id period tot_score
10011 1 3
10011 2 3
10011 3 2
10011 4 1
10022 1 4
10022 2 3
10022 3 4
10022 4 7
10031 1 1
10031 2 2
10031 3 2
10031 4 6
10032 1 6
10032 2 7
10032 3 4
10032 4 5
end
tempfile a
sort id period
save `a', replace
statsby _b[period] _se[period] e(df_r), ///
clear by(id): reg tot_score period
gen p = 2*ttail(_stat_3, abs(_stat_1/_stat_2))
gen byte neg = _stat_1 < 0 & p < .1
gen byte zero = p>.1
gen byte pos = _stat_1 > 0 & p < .1
keep id neg zero pos
sort id
merge id using `a'
assert _merge == 3
drop _merge
sort id period
list, sepby(id)
*------------------ end example ----------------------
( For more on how to use examples I sent to statalist see:
http://www.maartenbuis.nl/stata/exampleFAQ.html )
Hope this helps,
Maarten
--------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany
http://www.maartenbuis.nl
--------------------------
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
