Dear Enzo
Coviello,
I believe
you are the one who co-wrote stcompet
which provides analytical possibilities crucial in studies with along follow up
and competing risks, resulting in cumulative probabilities that could be quite
different from those obtained ignoring the existence of competing risks.
In the last
North American Stata users meeting in Boston,
�Mr. Guttierez who represented STATA seemed to include my proposal to complete the
routine, by adding adjustment for covariates, in the tasks that stata will
pursue. Or so it seemed
I have
not heard from him since nor noticed that anything.
Do you
know of any plan to advance this matter? This has enormous importance in analysis
of the old cohort studies where long-term follow up of mortality and/or the incidence
of dementia have been instituted and relating them to mid-life levels of risk
factors is sought.
Many thanks
for your attention,
Uri
Goldbourt
Department
of Epidemiology and Preventive medicine
Tel
Aviv University
-----Original Message-----
From: owner-[email protected]
[mailto:owner-[email protected]]
On Behalf Of Enzo Coviello
Sent: Monday, March 10, 2008 5:50 PM
To: [email protected]
Subject: Re: st: simulation of survival datasets in Stata
I would add to the very useful reply of Paul a
reference whose reading my be of some practical help:
-Generating survival times
to simulate Cox proportional hazards model-
Stat in Med 2005; 24: 1713-1723
Best wishes.
Enzo
At 13.20 10/03/2008, you wrote:
>You need to simulate
time-to-event followed by
>censoring.� The
easiest is exponential or Weibull.
>Weibull is :
>S=exp(-lambda*t^p)
>where lambda is the
hazard ratio:
>lambda=exp(sum beta x)
>
>beta_0 is a constant
reflecting the baseline
>hazard.� beta_1 ... reflect
the effects of covariates.
>p is the Weibull shape
parameter (positive
>number), and if valued
p=1 reduces to exponential.
>
>Generate lambda with
hazard ratio 1.9 for drug treatment (0,1):
>
>- lambda=-2.5+ln(1.9)*drug -
>
>To simulate time-to-event,
substitute the random
>unit variate R for S, and
solve for t.
>t=(-ln(R)/lambda)^(1/p)
>
>or in Stata:
>
>- gen t=(-ln(uniform())/lambda)^(1/p) -
>
>Some of the t values
will be very large, and you
>will need a censoring
variable to reflect the end of study (say 90 days).
>- gen fail=1 -
>- replace fail=0 if t>90 -
>- replace t=90 if t>90 -
>
>
>Then, look at your
handiwork-
>- stset t, fail(fail) -
>- sts graph, by(drug) -
>- stcox drug -
>
>Paul
>
>
>
>
>
>
>
>Michael McCulloch
wrote:
>>Hello Statalist
members,
>>I'm seeking
examples of simulation of survival
>>datasets in Stata.
Can anyone point me in the
>>right direction? My
goal is to simulate a
>>dataset to
validate a Stata survival program I've written.
>>
>>Thank you.
>
>--
>E. Paul Wileyto, Ph.D.
>Assistant Professor of
Biostatistics
>Tobacco Use Research
Center
>School of Medicine, U.
of Pennsylvania
>3535 Market Street, Suite
4100
>Philadelphia, PA� 19104-3309
>
>215-746-7147
>Fax: 215-746-7140
>[email protected]
>http://mail.med.upenn.edu/~epw/
>*
>*�� For searches and
help try:
>*�� http://www.stata.com/support/faqs/res/findit.html
>*�� http://www.stata.com/support/statalist/faq
>*�� http://www.ats.ucla.edu/stat/stata/
Enzo Coviello
Unit� di Epidemiologia e Statistica ASL BARI
Piazza V.� Emanuele 14
70054 Giovinazzo (Bari)
Italy
tel./fax +39 080 3357867
mobile +39 347 5016 016
home +39 0883 695055
*
*�� For searches and help
try:
*�� http://www.stata.com/support/faqs/res/findit.html
*�� http://www.stata.com/support/statalist/faq
*�� http://www.ats.ucla.edu/stat/stata/
