Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Bonferroni-holm


From   "Roger B. Newson" <r.newson@imperial.ac.uk>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Bonferroni-holm
Date   Mon, 20 Aug 2012 13:51:39 +0100

The Bonferroni and Holm procedures, and other frequentist multiple-test procedures, are explained, and Stata implementations described, in Newson (2010) and Newson et al. (2003). The 2010 reference is more up to date, as it describes q-values, which most people nowadays view as an improvement on discovery sets. The q-value package is -qqvalue-, and the discovery-set package is -smileplot-. Both of these can be downloaded from SSC using the -ssc- command.

I hope this helps.

Best wishes

Roger


References

Newson RB. Frequentist q-values for multiple-test procedures. The Stata Journal 2010; 10(4): 568-584. Purchase from
http://www.stata-journal.com/article.html?article=st0209

Nwwson R. and The ALSPAC Study Team. Multiple-test procedures and smile plots. The Stata Journal 2003; 3(2): 109-132. Download from
http://www.stata-journal.com/article.html?article=st0035

Roger B Newson BSc MSc DPhil
Lecturer in Medical Statistics
Respiratory Epidemiology and Public Health Group
National Heart and Lung Institute
Imperial College London
Royal Brompton Campus
Room 33, Emmanuel Kaye Building
1B Manresa Road
London SW3 6LR
UNITED KINGDOM
Tel: +44 (0)20 7352 8121 ext 3381
Fax: +44 (0)20 7351 8322
Email: r.newson@imperial.ac.uk
Web page: http://www.imperial.ac.uk/nhli/r.newson/
Departmental Web page:
http://www1.imperial.ac.uk/medicine/about/divisions/nhli/respiration/popgenetics/reph/

Opinions expressed are those of the author, not of the institution.

On 19/08/2012 16:13, Sadjad Riahi wrote:
dear billy,

thank you very much for the advice. my part of analysis is not
voxel-wise but volumetric analysis so I have about 16 volumes to test.
Thank you for your advice, I will try what you said.
thank you very much again,
sadjad

On Sun, Aug 19, 2012 at 7:24 PM, William Buchanan
<william@williambuchanan.net> wrote:
Hi Sadjad,

While Tony had suggested reading a book on multiple comparisons, I'm not sure that would get you all the way to the answer that you need.  Depending on the voxel size that was used for your scan, it is possible to have upwards of 30,000 dependent variables; in this case to reach a significant finding using a typical Bonferroni correction you would need to observe a p value of less than 0.00000167.  I'm not aware of any user written programs in Stata that are used for neuroimaging analysis, so I would suggest you may have an easier time doing this by using software designed for this type of analysis.  Off the top of my head, I know there there are some modules in R for this, there are also a series of routines written in Python that are fairly common, and then there are commercially available packages in MatLab (among others).  If you do a search for ANFI (another freely distributed package) you can probably find a link that would direct you to many of the programs funded b
!
  y !
  NIH for the development of neuroimaging analyses.

Has your data already been cleaned/prepped?  If not that is an incredibly time consuming task that you should focus on first.

HTH,
Billy

Sent from my iPhone

On Aug 19, 2012, at 7:26, Sadjad Riahi <sadjad.riahi@gmail.com> wrote:

dear billy,
thank you for your kind reply.Please accept my apologies if the
question sounds silly but I am a medical student and newbie in the
field of research and this is my first project, I have searched hard
for this on the web but no luck.the original paper is: Rimol LM,
Hartberg CB, Nesvag R, et al. Cortical thickness and subcortical
volumes in schizophrenia and bipolar disorder. Biol Psychiatry
2010;68:41-50. it has done the analysis via SPSS which I know there is
no such correction option in SPSS.
yours,
Sadjad Riahi

On Sun, Aug 19, 2012 at 6:22 PM, William Buchanan
<william@williambuchanan.net> wrote:
Hi Sadjad,

As stated in the FAQ, you are instructed to provide full references to material when referencing anything.  Additionally, the correction methods used in neuroimaging analyses are likely to differ to some degree compared with those used in social sciences, since it is much less likely that someone in the social sciences would be working with several tens of thousands of dependent variables simultaneously.  Most software for neuroimaging analyses will have built in support for these corrections (at least Brain Voyager does).  It would be helpful to provide more explicit detail to the list if you would like any reasonable advice.

HTH,
Billy

Sent from my iPhone

On Aug 19, 2012, at 6:33, Sadjad Riahi <sadjad.riahi@gmail.com> wrote:

dear statalist,
I have found In a paper with the similar methodology with ours that
for correction of multiple comparisons, authors have applied
bonferroni-holm correction. To explain the analysis in short, they
have several dependent variable (subcorticalgray matter  region
volumes) and three groups of diagnosis, intracranial volumes,age and
sex as predictor variables.they have fitted a GLM for each region
volume as the dependant variable seperately, then F test was preformed
and subsequently contrast analysis. but for multiple comparison they
first corrected for the omnibus test and then correction for pairwise
comparison for each GLM that survived the first round of correction
was performed. Can you please enlighten me with this correction
procedure more in detail.

yours,

Sadjjad Riahi
*
*   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/

*
*   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/

*
*   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/

*
*   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/

*
*   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/

*
*   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/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index