Date:         Sat, 25 Jul 2009 12:16:33 -0700
Reply-To:     SR Millis <srmillis@yahoo.com>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         SR Millis <srmillis@yahoo.com>
Subject:      Re: Non-linear survival anlysis
Comments: To: kornbrot <d.e.kornbrot@herts.ac.uk>
Content-Type: text/plain; charset=utf-8

Have considered using restricted cubic splines?  Frank Harrell discusses this method in his book, Regression Modeling Strategies.

It can be easily implemented in R, which is free.




Scott R Millis, PhD, ABPP (CN,CL,RP), CStat, CSci
Professor & Director of Research
Dept of Physical Medicine & Rehabilitation
Dept of Emergency Medicine
Wayne State University School of Medicine
261 Mack Blvd
Detroit, MI 48201
Email:  smillis@med.wayne.edu
Tel: 313-993-8085
Fax: 313-966-7682


--- On Sat, 7/25/09, kornbrot <d.e.kornbrot@herts.ac.uk> wrote:

> From: kornbrot <d.e.kornbrot@herts.ac.uk>
> Subject: Non-linear survival anlysis
> To: SPSSX-L@LISTSERV.UGA.EDU
> Date: Saturday, July 25, 2009, 9:54 AM
>
>
> Non-linear survival anlysis
>
>
> Help
>
>
>
> I need to do NON_LINEAR cox regression, i.e. would like to
> know if the hazard function for death is  quadratic as
> well as a linear function of the covariate (quantity of
> Â a biomarker)
>
>
>
> There is no obvious way to do this using menus and spss
> didn’t like ne inserting a quadratic term in the
> syntax editor
>
> Obviously I can create variable that are x^2, x^3 etc of my
> predicotr covariate x – but there must be a better
> way, as for the usual logistic or regression function
>
>
>
> Best
>
> Diana
>
> Professor
> Diana Kornbrot
>
> email:Â  d.e.kornbrot@herts.ac.uk Â
> Â
>
> web:
> Â Â Â http://web.me.com/kornbrot/KornbrotHome.html
>
> Work
>
> School of Psychology
>
> Â University of Hertfordshire
>
> Â College Lane, Hatfield, Hertfordshire AL10 9AB, UK
>
>
>  Â voice:
> Â Â +44 (0) 170 728 4626
>
> Â Â Â fax: Â Â Â Â +44 (0) 170
> 728 5073
>
> Home
>
> Â 19 Elmhurst Avenue
>
> Â London N2 0LT, UK
>
> Â Â Â Â voice: Â Â +44 (0) 208 883
> Â 3657
>
> Â Â Â Â mobile: +44 (0) 796 890 2102
>
>
> Â Â Â fax: Â Â Â Â Â +44 (0)
> 870 706 4997
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>

=====================
To manage your subscription to SPSSX-L, send a message to
LISTSERV@LISTSERV.UGA.EDU (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD