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]

From |
Jacob Fowles <jacob.fowles@gmail.com> |

To |
statalist@hsphsun2.harvard.edu |

Subject |
Re: st: clogit:determining number of observations for each level of a categorical variable |

Date |
Mon, 27 Jun 2011 11:46:50 -0500 |

Daniel, I have used the user written command by Longton and Cox named "distinct" for this exact purpose ("ssc install distinct"). There is also a Stata FAQ on this subject that provides some other strategies: http://www.stata.com/support/faqs/data/distinct.html Hope this helps. Best, Jacob Jacob Fowles Assistant Professor Department of Public Administration University of Kansas 785-864-3527 jacob.fowles@ku.edu Jacob On Mon, Jun 27, 2011 at 9:36 AM, Daniel Herbert Opi <opi.herbert@gmail.com> wrote: > Dear Statalist, > I am new to Statalist and had a question that I hope I can get some > input on. I am carrying out conditional logistic regression-clogit > (example below) on a case control study where each case has been > matched to a control to look at the effect of several independent > categorical variables (xyz and abc in my example) on a dependent > variable of disease outcome (disease). The output in stata shows the > total number of observations used for the analysis (326 in this case) > but I was wondering whether there is a way of determining the number > of observations used for each level of the independent categorical > variables (xyz and abc) since I can already tell some observations (65 > in this case) have been dropped because of having all positive or > negative outcomes. > > . clogit disease i.xyz i.abc, strata (set1) or > note: 65 groups (65 obs) dropped because of all positive or > all negative outcomes. > Iteration 0: log likelihood = -107.72902 > Iteration 1: log likelihood = -107.64854 > Iteration 2: log likelihood = -107.64852 > Iteration 3: log likelihood = -107.64852 > Conditional (fixed-effects) logistic regression Number of obs = 326 > LR chi2(4) = 10.67 > Prob > chi2 = 0.0305 > Log likelihood = -107.64852 Pseudo R2 = 0.0472 > ------------------------------------------------------------------------------ > disease | Odds Ratio Std. Err. z P>|z| [95% Conf. Interval] > -------------+---------------------------------------------------------------- > xyz | > 1 | 2.380791 .951431 2.17 0.030 1.08782 5.210571 > 2 | 2.73225 1.09711 2.50 0.012 1.243735 6.002232 > | > abc | > 1 | .5346201 .1450661 -2.31 0.021 .3141063 .9099424 > 2 | .9274642 .3857671 -0.18 0.856 .4104408 2.095771 > ------------------------------------------------------------------------------ > > Regards > > Daniel > > * > * 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/

**Follow-Ups**:**RE: st: clogit:determining number of observations for each level of a categorical variable***From:*Nick Cox <n.j.cox@durham.ac.uk>

**References**:**st: clogit:determining number of observations for each level of a categorical variable***From:*Daniel Herbert Opi <opi.herbert@gmail.com>

- Prev by Date:
**st: RE: Case control study** - Next by Date:
**st: Link for dropmiss.ado** - Previous by thread:
**st: clogit:determining number of observations for each level of a categorical variable** - Next by thread:
**RE: st: clogit:determining number of observations for each level of a categorical variable** - Index(es):