Back to the Top
Dear Members,
greetings.
I have some problems regarding outliers in bioequivalence studies. I
have considered the book on bioequivalence "Design and Analysis of
Bioavailability and Bioequivalence " by Shein - Chung Chow,different
methods of outlier detection are explained here , on of them is
Hotelling T^2 distribution method , in this method the formula of
hotelling T^2 statistics is given but I m not understanding how can I
calculate the matrix of sum of squares and cross product which is used
in this formula
I am grateful to here your suggestion and thougths.
Regards.
Rasheed.
Back to the Top
The following message was posted to: PharmPK
Rasheed
The Hotelling T^2 test is just an extension of the t-test but for
multiple variables. Most statistical packages (including SPSS, NCSS
and many other acronymic software) and others have this as a standard
test with good documentation on how to calculate the test statistic.
Steve
Back to the Top
To do this calculation you need to either have a program like SAS
calculate it directly or have access to a matrix based programming
language like R, Matlab, Proc IML in SAS, or S-plus.
pete bonate
Back to the Top
Hi Mr. Rasheed,
You can Calculate Mahanalobis distance using SPSS and then using the
relation between Mahanalobis distance and Hotelling T^2 you can
directly calculate Hotelling T^2 OR you can also use SAS for direct
calculation of Hotelling T^2.
Thanks & Regards,
Ravi Pandey
Want to post a follow-up message on this topic?
If this link does not work with your browser send a follow-up message to PharmPK@boomer.org with "Hotelling T^2 distribution method" as the subject | Support PharmPK by using the |
Copyright 1995-2011 David W. A. Bourne (david@boomer.org)