[R] p-value=0 running coxph
Thomas Lumley
tlumley at u.washington.edu
Thu Oct 30 21:44:31 CET 2008
On Thu, 30 Oct 2008, GSt wrote:
> Dear all,
>I have a question concerning the p-value. When running coxph I get a p-value
>= 0. :confused:
>Can this be true? Why arenÿÿt there decimal points? Is there a way to
>find out the exact p-value?
The p-values are rounded, and of course they are not exactly zero, just
<0.0005. You are given the z-statistic so you could compute the p-value
> 2*pnorm(-9.16)
[1] 5.189732e-20
Now, this number is probably pretty useless because the Normal
approximation to the test statistic doesn't work that far out in the tails
unless you have a very large number of events. It's also unusual for the
null hypothesis to be even slightly interesting when the p-value is that
small -- do you really not know whether albumin concentrations are related
to the outcome?
-thoams
Here is the output:
----------------------------------------------------------------------------------------------------
Call:
coxph(formula = Surv(start, stop, status) ~ Albumin_gproL, data = dial,
na.action = na.omit, method = "breslow")
n=22449 (31 observations deleted due to missingness)
coef exp(coef) se(coef) z p
Albumin_gproL -0.157 0.855 0.0172 -9.16 0
exp(coef) exp(-coef) lower .95 upper .95
Albumin_gproL 0.855 1.17 0.826 0.884
Rsquare= 0.003 (max possible= 0.033 )
Likelihood ratio test = 73.6 on 1 df, p=0
Wald test = 83.8 on 1 df, p=0
Score (logrank) test = 86.1 on 1 df, p=0
> cox.zph(cox_1111)
rho chisq p
Albumin_gproL -0.0509 0.197 0.657
----------------------------------------------------------------------------------------------------
Thank you in advance...
GSt
--
View this message in context: http://www.nabble.com/p-value%3D0-running-coxph-tp20248274p20248274.html
Sent from the R help mailing list archive at Nabble.com.
______________________________________________
R-help at 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.
Thomas Lumley Assoc. Professor, Biostatistics
tlumley at u.washington.edu University of Washington, Seattle
More information about the R-help
mailing list