Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: RE: PKSHAPE bug?


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: PKSHAPE bug?
Date   Wed, 5 Dec 2007 19:59:42 -0000

Your report here boils down to "figures appear incorrect" without either

_complete_ code or _complete_ examples with both data and results.
 
I suggest you follow the guidelines at 
http://www.stata.com/support/faqs/techsup/sendout.html 
and liaise directly with Stata tech-support; otherwise there 
is unlikely to be enough for anyone to work on here. 

Nick
n.j.cox@durham.ac.uk 

Roberto Melotti

I have been using PKSHAPE to reshape data for a 3X3 crossover trial.
I have data originally arranged by period for the 6 periods of the trial
(i.e. 1=baseline, 2=period 2, 3=1st wash-out, etc.)
I have seq (variable for sequence) coded as string of the type '0a0b0c'
 
I have pk-reshaped the dataset to obtain a long format dataset in order
to be used by the command PKCROSS.
namely I typed: pkshape id seq phv1 phv2 phv3 phv4 phv5 phv6,
outcome(phv) 
 
I have saved the results in a new dataset and repeated the same command
for another outcome (phn); saved results in another dataset as well.
 
I have then merged together the two datasets matching by id and
sequence.
 
I have therefore run the command pkcross on each outcome variable
separately and everything seemed to work fine (the PKCROSS output is
straightforward).
 
Nonetheless when I tried to calculate the averages and standard
deviations for each outcome by treatment (with the command table), the
figures appear incorrect!
 
I have checked this with the averages on the outcome variable by period,
sequence and in such instance the problem doesn't occur: the estimates
are corrrect!
 
Has anyone else experienced the same problem?
 

*
*   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/



© Copyright 1996–2022 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index