Many thanks for your question that I did not dare asking
and for the instructive responses.
Another example of plotting hazard can be found in
Retsky M et al, breast cancer research:
http://breast-cancer-research.com/content/6/4/R372
Regarding plot of hazard ratio across time,
perhaps most convenient is:
somefittedcoxmodel <- coxph(yourmodel)
plot(cox.zph (somefittedcoxmodel, transform='identity'))
Regards,
Vincent Vinh-Hung
-----Original Message-----
From: s-news-owner@lists.biostat.wustl.edu
[mailto:s-news-owner@lists.biostat.wustl.edu] On Behalf Of Maggie Cheang
Sent: Wednesday, September 28, 2005 10:55 AM
To: Prof Brian Ripley
Cc: Terry Therneau; s-news@lists.biostat.wustl.edu
Subject: Re: [S] instant hazard rate/survival
I really appreciate your advice. I am a beginner in survival
analysis, in fact, in S/R.
Maybe i should clarify what I would like to do.
I would like to look to plot both the instant hazard rate of a
biomarker positive and negative expression respectively, and also the
hazard ratio of the positive in respect to the negative expression,
adjusting with other factors, such as tumor size and age.
I read this reference paper (Estrogen receptors and distinct patterns
of breast cancer relapse, by Kenneth R. Hess et al, Breast cancer
research and treatment 78: 105-118, 2003). In the paper, the authors
plot the a smooth hazard graph, and also hazard ratio.
I am not quite sure how the author draw the smooth hazard ratio, but
I like the idea that plotting the hazard ratio across time, to
visualize the changes.
Thanks in advance.
Maggie
On 28-Sep-05, at 1:18 AM, Prof Brian Ripley wrote:
> On Tue, 27 Sep 2005, Maggie Cheang wrote:
>
>
>> Thanks for Terry to reply for my qustion.
>>
>
> Note that it is implicit in Terry's answer that the hazard is smooth.
> You have to make some asssumption here, and his includes how smooth
> it is.
>
>
>> That comes to my second question,
>> is there a simple way to plot the Hazard ratio from a coxph?
>>
>
> What hazard ratio is that? Note the theory underlying coxph does
> not assume that hazards are smooth but that they are very rough.
>
>
>> anyone familiar with the muhaz package?
>>
>
> Yes, and it is discussed (alongside similar ideas) in the MASS on-
> line complements. See http://www.stats.ox.ac.uk/pub/MASS4/
> #Complements .
>
>
>
>> thanks a bunch
>> Maggie
>> On Tuesday, September 27, 2005, at 06:24AM, Terry Therneau
>> <therneau@mayo.edu> wrote:
>>
>>
>>> The simplest way that I know to get an estimate of the
>>> hazard is to use smoothing splines.
>>>
>>> fit <- survfit(Surv(time, status) ~ sex, data=lung)
>>>
>>> temp1 <- smooth.spline(fit[1]$time, 1-fit[1]$surv, df=5)
>>> temp2 <- smooth.spline(fit[2]$time, 1-fit[2]$surv, df=5)
>>>
>>> plot( predict(temp1, deriv=1), type='l')
>>> lines(predict(temp2, deriv=1), col=2)
>>>
>>>
>>> ------
>>> The lung cancer data set is part of Splus (it's used in several
>>> manual pages as an example). The plot shows the males to have a
>>> higher initial hazard than the females.
>>>
>>> You choice of 5 degrees of freedom for the spline was completely
>>> arbitrary.
>>>
>>> Terry Therneau
>>>
>>> --------------------------------------------------------------------
>>> This message was distributed by s-news@lists.biostat.wustl.edu. To
>>> unsubscribe send e-mail to s-news-request@lists.biostat.wustl.edu
>>> with
>>> the BODY of the message: unsubscribe s-news
>>>
>>>
>>>
>> --------------------------------------------------------------------
>> This message was distributed by s-news@lists.biostat.wustl.edu. To
>> unsubscribe send e-mail to s-news-request@lists.biostat.wustl.edu
>> with
>> the BODY of the message: unsubscribe s-news
>>
>>
>
> --
> Brian D. Ripley, ripley@stats.ox.ac.uk
> 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, UK Fax: +44 1865 272595
> --------------------------------------------------------------------
> This message was distributed by s-news@lists.biostat.wustl.edu. To
> unsubscribe send e-mail to s-news-request@lists.biostat.wustl.edu with
> the BODY of the message: unsubscribe s-news
>
Maggie Cheang
mcheang@mac.com
--------------------------------------------------------------------
This message was distributed by s-news@lists.biostat.wustl.edu. To
unsubscribe send e-mail to s-news-request@lists.biostat.wustl.edu with
the BODY of the message: unsubscribe s-news
|