home *** CD-ROM | disk | FTP | other *** search
- Comments: Gated by NETNEWS@AUVM.AMERICAN.EDU
- Path: sparky!uunet!zaphod.mps.ohio-state.edu!sdd.hp.com!spool.mu.edu!darwin.sura.net!paladin.american.edu!auvm!MAIL.SWIP.NET!MASTFOH
- Message-ID: <199212141435.AA06865@mail.swip.net>
- Newsgroups: bit.listserv.stat-l
- Date: Mon, 14 Dec 1992 15:34:41 -0100
- Sender: STATISTICAL CONSULTING <STAT-L@MCGILL1.BITNET>
- From: MAGNUS STENBECK <mastfoh%tellus.sos.se%sos@MAIL.SWIP.NET>
- Subject: PC software for proportional hazards model
- Lines: 24
-
- I am analyzing survival time for several major diagnostic groups of
- cancer. The data is from the Swedish cancer registry. The file
- contains some 1 million observations - all malignant cancers
- discovered in Sweden since 1958 are included in the material.
- I am analyzing the material using a 486-50 Mhz PC using a
- regression approach, i.e. logit analyses on the
- likelihood of surviving 1, 5, or 10 years, and censored exponential
- regression models predicting survival time. I also want to use
- proportional (and possibly non-proportional) hazards models. I
- normally work with GAUSS which is fast enough to do analyses of up
- to 100,000 cases at a time, but I cannot find applications written
- in GAUSS for the hazards models. Does anyone know of some good
- software for the micro computer which does the proportional hazards
- model and which is fast enough to handle, say, 5 explanatory
- variables and 100,000 observations?
- (SAS, for instance, is not useful because of the excessive time SAS
- takes to analyze 50-100,000 cases on a micro computer ....)
-
- Magnus Stenbeck Email: MASTFOH at TELLUS.SOS.SE
- Center for Epidemiology
- National Board of Health and Welfare
- Banergatan 16
- S-106 30 Stockholm
- SWEDEN
-