[R] help: how to get coefficients from the result
mingan
mingan at math.unm.edu
Thu Feb 24 23:11:56 CET 2005
I am using timereg and Survreg packages
below is the output, but how can I get the coefficient for zt1 zt2
directly,
that is how can I get 0.000946 and 0.963000 directly ?
I tried out$coeff[1,1] or sth. similar, but does not work
since I will run a simulation of several thousand, so I should get the
coefficients directly
thanks
> out=prop.odds(Surv(t,d)~zt1+zt2)
Proportional odds model
> summary(out)
Proportional Odds model
Test for baseline
Test for non-siginificant effects
sup| hat B(t)/SD(t) | p-value H_0: B(t)=0
Baseline 11.9 0
Test for time invariant effects
sup| B(t) - (t/tau)B(tau)| p-value H_0: B(t)=b t
Baseline 358 0.342
Covariate effects
Coef. Std. Error Robust SE D2log(L)^-1
zt1 0.000946 0.0890 0.0885 0.0891
zt2 0.963000 0.0574 0.0584 0.0541
Score Tests for Goodness-of-fit
sup| hat U(t) | p-value H_0
zt1 6.73 0.848
zt2 12.20 0.798
Call:
prop.odds(Surv(t, d) ~ zt1 + zt2)
n=50; iter=1000
beta1=0;beta2=1
zt1=NULL
zt2=NULL
t=NULL
for( i in 1:n)
{
z1=rbinom(1,1,0.5)
z2=rexp(1,1)
zt1=c(zt1,z1)
zt2=c(zt2,z2)
u=runif(1,0,1)
temp=exp(-z1*beta1-z2*beta2)*u/(1-u)
t=c(t,temp)
}
d=rep(1,n)
mt=cbind(t,d,zt1,zt2)
out=prop.odds(Surv(t,d)~zt1+zt2)
summary(out)
More information about the R-help
mailing list