Statalist


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

st: RE: problem with outreg2: p-value deciles


From   "Newson, Roger B" <r.newson@imperial.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: problem with outreg2: p-value deciles
Date   Thu, 14 Feb 2008 16:09:17 -0000

I don't use -outreg2- myself )because I use -parmest-). However, my personal view re formatting of P-value is that they should be reported to a given number of significant figures, rather than to a given number of decimal places. This is because (contrary to popular belief) P-values below 0.0005 are NOT "as good as zero", because today's genome-scan chips can measure a million single-nucleotide polymorphisms (SNPs), and today's computers can measure the associations of all of these polymorphisms with a disease, and over 1/2000 of these associations (on average) will be statistically significant at P<=0.0005, even if all associations are zero in the population at large. Therefore, it matters whether an individual P-value is 0.0005 or 0.00000005.

The ideal Stata format to use for P-values is probably %-8.1e, ie left-justified, with a width of 8, presented in scientific notation as x.ye-z, where x and y are digits, and z is a possibly multi-digit integer, which, in practice, will not have more than 3 digits, even on today's computers (I think). However, as a lot of readers are more accustomed to P-values of 0.05 than to P-values of 5.0e-2, we might compromise with these readers by choosing a format of %-8.1g, which should give the familiar multi-digit decimal format for P-values down to, and including, 1.0e-5.

I would be surprised if -outreg- and/or -outreg2- and/or -estaou- did not have the ability to format P-values in either of these ways. (Or in a format such as %8.3f, if readers ostinately prefer decimal-place formats.) The -parmest- package has a -format()- option for this purpose.

I hope this helps.

Best wishes

Roger


Roger B Newson
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: 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.

-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of John Bunge
Sent: 14 February 2008 15:23
To: statalist@hsphsun2.harvard.edu
Subject: st: problem with outreg2: p-value deciles

Dear all,

I am using outreg2 with the 'pvalue' option (p-values of t-statistics instead of standard errors are reported).

I tried to limit the number of deciles of the p-values to 3 with the 'tdec(integer)' option but it did not work. For some coefficients, it reports me 8 deciles (like e.g. 0.00000071) which futilely blows up the width of the table columns. Instead of this space-consuming number I'd prefer a simple 0.000. 

Any helpful suggestions are highly appreciated.

Best,

John.
_______________________________________________________________________
Jetzt neu! Schützen Sie Ihren PC mit McAfee und WEB.DE. 30 Tage
kostenlos testen. http://www.pc-sicherheit.web.de/startseite/?mc=022220


*
*   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/

*
*   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 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index