[R] Projecting Survival Curve into the Future
Hello, I have a survivor curve that shows account cancellations during the past 3.5 months. Â Fortunately for our business, but unfortunately for my analysis, the survivor curve doesn't yet pass through 50%. Â Is there a safe way to extend the survivor curve and estimate at what time we'll hit 50%? We started a new program 3.5 months ago, and I believe that this set of accounts behaves differently than the rest of our company's accounts. Thanks very much, Alan -- Alan Cox Director, User Experience iContact, Corp. p 919.459.1038 f 919.287.2475 [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Projecting Survival Curve into the Future
I have a survivor curve that shows account cancellations during the past 3.5 months. Â Fortunately for our business, but unfortunately for my analysis, the survivor curve doesn't yet pass through 50%. Â Is there a safe way to extend the survivor curve and estimate at what time we'll hit 50%? Without any example code it's hard to say, but take a look at ?predict.coxph and ?predict.survreg in the survival package. Regards, Richie. Mathematical Sciences Unit HSL ATTENTION: This message contains privileged and confidential inform...{{dropped:20}} __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Projecting Survival Curve into the Future
You might consider a probit analysis using ln(Time) as the dose. At 09:24 AM 9/4/2008, [EMAIL PROTECTED] wrote: I have a survivor curve that shows account cancellations during the past 3.5 months. Â Fortunately for our business, but unfortunately for my analysis, the survivor curve doesn't yet pass through 50%. Â Is there a safe way to extend the survivor curve and estimate at what time we'll hit 50%? Without any example code it's hard to say, but take a look at ?predict.coxph and ?predict.survreg in the survival package. Regards, Richie. Mathematical Sciences Unit HSL Robert A. LaBudde, PhD, PAS, Dpl. ACAFS e-mail: [EMAIL PROTECTED] Least Cost Formulations, Ltd.URL: http://lcfltd.com/ 824 Timberlake Drive Tel: 757-467-0954 Virginia Beach, VA 23464-3239Fax: 757-467-2947 Vere scire est per causas scire __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Projecting Survival Curve into the Future
On Thu, 4 Sep 2008, [EMAIL PROTECTED] wrote: I have a survivor curve that shows account cancellations during the past 3.5 months. Â Fortunately for our business, but unfortunately for my analysis, the survivor curve doesn't yet pass through 50%. Â Is there a safe way to extend the survivor curve and estimate at what time we'll hit 50%? Without any example code it's hard to say, but take a look at ?predict.coxph and ?predict.survreg in the survival package. You will not be able to do this with coxph: there will be no events to estimate the baseline hazard from. Whether using a parametric accelerated life model (survreg) is 'safe' depends on what you are prepared to assume. I'd say it was pretty dubious unless you have theoretical reasons to suppose that e.g. an exponential is a good approximation and it fits the data you do have. Regards, Richie. -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595__ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.