Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
st: pkequiv and p-values from one-sided t-tests
From
Nigel Perkins <[email protected]>
To
[email protected]
Subject
st: pkequiv and p-values from one-sided t-tests
Date
Mon, 25 Feb 2013 16:28:11 +1000
I am using pkequiv in Stata 12.1 to analyse data for bioequivalence
. pkequiv value treat period sequence id, tost
Classic confidence interval for bioequivalence
----------------------------------------------------------------------
[equivalence limits] [ test limits ]
----------------------------------------------------------------------
difference: -86.775 86.775 -150.067 185.817
ratio: 80% 120% 65.412% 142.827%
----------------------------------------------------------------------
probability test limits are within equivalence limits = 0.0160
Schuirmann's two one-sided tests
----------------------------------------------------------------------
upper test statistic = -0.797 p-value = 0.456
lower test statistic = 1.211 p-value = 0.271
Everything makes sense to me except the reported p-values for the
Schuirmann's tests.
The DF for this dataset is equal to 6 based on a 2-period cross-over study
with n=4 in each sequence.
The reported p-values appear to be two-sided p-values and not one-sided
p-values. I can estimate one-sided p-values using Excel or a web-calculator
and they are numerically equal to half the p-values reported by pkequiv. I
expected the Schuirmann tests to each report a one-sided p-value.
I have repeated this analysis using a couple of different datasets and the
results appear to be consistent.
I am wondering if I am doing something wrong here. Any comment would be much
appreciated.
Nigel Perkins
AusVet, Australia
[email protected]
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/