Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | "Roger B. Newson" <r.newson@imperial.ac.uk> |
To | statalist@hsphsun2.harvard.edu |
Subject | Re: st: Cross-validation of Harrell's C statistics |
Date | Thu, 02 Aug 2012 10:41:11 +0100 |
Best wishes Roger 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 02/08/2012 07:57, Jonas Bille Nielsen wrote:
Thank you very much for your replay. I have already read your article and used your excellent codes for calculating C statistics based on splitting the dataset into a training and a test set. However, I prefer to do the cross-validation with the use a bootstrap-like method with repeated splitting of the dataset as you also suggest in section 5 of your paper. Have you done such a procedure yourself, again with the use of somersd? Thanks, Jonas On Wed, Aug 1, 2012 at 7:31 PM, Roger B. Newson <r.newson@imperial.ac.uk> wrote:U haven't written a script for this whole procedure. However, I have written a Stata Journal article about how to carry out such a procedure in Stata using -somersd-, and the issues involved (Newson, 2010). Best wishes Roger References Newson RB. Comparing the predictive power of survival models using Harrell’s c or Somers’ D. The Stata Journal 2010; 10(3): 339–358. Purchase from http://www.stata-journal.com/article.html?article=st0198 or download pre-publication draft drom http://www.imperial.ac.uk/nhli/r.newson/papers.htm 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 01/08/2012 16:55, Jonas Bille Nielsen wrote:Dear statalisters I'm doing Cox models and want to compare a model 'A' with a model 'B' with regards to Harrell's C statistics within the same dataset and with the use of a statistical test. As I'm calculating C statistics in the same dataset in which the Cox models were fitted, I will have an over-optimistic estimate of C. I want to take this overoptimism into account by some kind of cross-validation. As my dataset is very large, I can't use the rather slow 'estat concordance' command, but has to use the 'somersd' command or something as fast at this command. Do someone have a script for such a procedure? Thanks in advance,* * 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/-- Jonas Bille Nielsen, MD Laboratory for Molecular Cardiology, 9312 Copenhagen University Hospital, Rigshospitalet Blegdamsvej 9, DK-2100 jonas.bille.nielsen@gmail.com Work (+45) 35 45 65 06 Mobile (+45) 40 89 20 90 * * 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/