BookmarkSubscribeRSS Feed
coralie
Calcite | Level 5

Hello everyone,

I did a proc lifereg using the generalized gamma distribution, as follow :

proc lifereg data=survival.data;

      class treatment;

      model timedays*death(0)=treatment/dist=gamma;

run;

I want to identify the 3 parameters of the generalized gamma distribution. I know this distribution has one scale parameter and two scales parameters, but how can I calculate them using the output ?

Here is what I get in the output :

                                      The LIFEREG Procedure

                       Analysis of Maximum Likelihood Parameter Estimates

                                   Standard   95% Confidence     Chi-

            Parameter       DF Estimate    Error       Limits       Square Pr > ChiSq

            Intercept        1   7.3382   0.0799   7.1815   7.4949 8425.03     <.0001

            treatment     0  0   0.0000    .        .        .         .        .

            Scale            1   0.4830   0.3039   0.1407   1.6581

            Shape            1   1.5785   1.0907  -0.5593   3.7163

Thank you very much for your help.

3 REPLIES 3
RodS
Calcite | Level 5

I'm having a similar issue. In particular, given the parameterization as in the file linked below, I want to calculate the expected value, but I can't work out the values of alpha, beta, and gamma from the intercept, scale, and shape estimates.

 

PDF and E(X) 

 

Thanks!

 

Rick_SAS
SAS Super FREQ

I don't know the answer without research, but I recall that Joseph Gardiner gave a 2012 paper about PROC LIFEREG in which he discusses the various distributions. See

"Modeling heavy-tailed distributions in healthcare utilization by parametric and Bayesian methods"

 

HTH.

RodS
Calcite | Level 5

Hi, thanks for the reply and the reference. Unfortunately, in his table of distributions he includes the PDF and expected value formulae only for the common two-parameter gamma distribution.

 

Best,

Rod

sas-innovate-2024.png

Join us for SAS Innovate April 16-19 at the Aria in Las Vegas. Bring the team and save big with our group pricing for a limited time only.

Pre-conference courses and tutorials are filling up fast and are always a sellout. Register today to reserve your seat.

 

Register now!

What is ANOVA?

ANOVA, or Analysis Of Variance, is used to compare the averages or means of two or more populations to better understand how they differ. Watch this tutorial for more.

Find more tutorials on the SAS Users YouTube channel.

Discussion stats
  • 3 replies
  • 2192 views
  • 0 likes
  • 3 in conversation