Proc Lifereg Hazard

Similar Threads:

1.Difference between proc reg and proc lifereg.

Hi,

What is the difference between a proc reg and  proc lifereg procedure. If
possible i need an explanation to be not too
technical.

Thanks,
Sa

2.PROC LIFEREG

Hello,

My problem is the following : I'm working with the PROC LIFEREG and I
would like to calculate the estimated time on validation data.
Does anyone know which formula I have to apply on my validation data
to get the estimated time ?

I used the Exponential distance.

proc LIFEREG data=training outest=A;
model TIME*cens(1)=&VARLIST/DIST=EXPONENTIAL;
output out=B xbeta=XB ;
run;

data predict;
 set validation;
estimate_time=exp(xbeta); /*I don't find the exact formula to apply*/
run;


Thanks in advance,

Please send me your ideas to my office mail :
 XXXX@XXXXX.COM 

Nathalie

3.proc lifereg generalized gamma distribution

When using PROC LIFEREG, what does the coefficient reported for covariates
mean? Is it the hazard ratio, or does the hazard ratio have to be
calculated using the scale parameter in the same way as when using the
Weibull distribution in SAS?

Jamie Griffin.

4.PROC LIFEREG: Estimating Failure Time Distribution

Hi All,

The problem is as follows: I have a set of customers, who took their
first product on a particular month of a year. I have observed them over
a 12 month window to check if they have taken up a second product. My
problem is to find out, if the customer will take up a second product
and when.

While using PROC LIFEREG, the failure time distribution (Option DIST = blah
blah in PROC LIFEREG) I am estimating is not correct. Hence my predicted
survival probabilites are going haywire. I have tried out all the "DIST = "
options provided in SAS.

Is there anyway I can force fit the underlying failure time distribution
given that, I have the hazard function and survival function available to
me from the Life Tables using PROC LIFETEST.

Awaiting your answers.

Thanks
Nirmalya

5.PROC LIFEREG for Tobit model

I tried to replicate the Tobit Regression in Long (1997:p.189). At the
begining, I generated population data with around 200 observations
following y=1.2+0.08x+e. Then, I censored observation if y<1.
In theory, the tobit regression should have to generate the similar
coefficents of the population data. However, I could not get it.
Here are my SAS code. Do anybody have a good idea about this.
Many thanks in advance.

DATA test;
DO I=0 to 14 by 0.07;
	X=i;
	Y=1.2+0.08*X+RANNOR(0);
	OUTPUT;
END;
run;
DATA test2;
set test;
IF y<1 then censor=1;
ELSE censor=0;
IF censor=1 then Y=0;
if y eq 0
         then Lower=.;
         else Lower=y;
run;
PROC LIFEREG data=test2;
 model (lower,y)=x/d=normal;
run;

6. What do the Estimates from PROC LIFEREG Represent?

7. Using Proc NLMIXED to estimate a multilevel discrete-time hazard model

8. AW: phreg proc&proportional hazard assumption