Statalist


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

st: PKSHAPE bug?


From   Roberto Melotti <[email protected]>
To   "'[email protected]'" <[email protected]>
Subject   st: PKSHAPE bug?
Date   Wed, 5 Dec 2007 17:35:23 -0000

Dear All,
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?
 
Any help would be highly appreciated,
 
Regards, Roberto.
 

***********************************

Roberto Melotti, Senior Statistician

DISCLAIMER: The information in this message is confidential and may be legally privileged. It is intended solely for the addressee. Access to this message by anyone else is unauthorised. If you are not the intended recipient, any disclosure, copying, or distribution of the message, or any action or omission taken by you in reliance on it, is prohibited and may be unlawful. Please immediately contact the sender if you have received this message in error. Thank you.






© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index