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]

From |
Nick Cox <n.j.cox@durham.ac.uk> |

To |
"'statalist@hsphsun2.harvard.edu'" <statalist@hsphsun2.harvard.edu> |

Subject |
st: RE: xtile command for calculating tertiles |

Date |
Sun, 24 Oct 2010 18:01:55 +0100 |

Answering this well is possible only to a very small number of people using both Stata and SAS; I am not among them. You'd get a good answer faster with either or both of 1. Results from the same test data containing ties administered to both programs. For example, the numbers 1/29 30 repeated 10 times 40/100. 2. An explanation of what exactly SAS does, in this case, gleaned by looking at its code or its documentation. There is a literature explaining that different programs use one or more of about ten different recipes for quantiles, for example Hyndman, R.J. and Y. Fan. 1996. Sample quantiles in statistical packages. The American Statistician 50: 361--365. Nick n.j.cox@durham.ac.uk Weichle, Thomas Can anyone comment on the difference between the way Stata's -xtile- command creates tertiles compared to the way the SAS -proc rank- creates tertiles? And the differences in which ties are handled? The following are the code I'm using and I am getting slightly different results. My data do contain ties and I think this is the issue. Stata: xtile tertile = opcd, nquantiles(3) SAS: proc rank data = test out = ranks groups = 3; var opcd; ranks opcd_tert; run; * * 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/

**References**:**st: xtile command for calculating tertiles***From:*"Weichle, Thomas" <Thomas.Weichle@va.gov>

- Prev by Date:
**RE: st: How does Stata calculate percentiles?** - Next by Date:
**RE: st: RE: Venn diagrams** - Previous by thread:
**st: xtile command for calculating tertiles** - Next by thread:
**re: st: RE: Re: standardized betas in Prais-Winsten Regression** - Index(es):