Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: parametric survival

From (Roberto G. Gutierrez, StataCorp)
Subject   Re: st: parametric survival
Date   Fri, 01 Apr 2011 11:09:20 -0500

Farzaneh Amanpour <> writes:

> I am fitting the generalized gamma distribution to survival data.I need
> all three parameters of generalized gamma to have linear relationship
> with covariates without having the intercept.The command
> "streg[varlist],dist(gamma) anc(varlist) anc2(varlist) noconstant" only
> model the first parameter without intercept, and would not remove the
> constant terms in two ancillary parameters. What can I do to suppress
> the constant terms in two ancillary parameters('anc' and 'anc2')?

Steven J. Samuels <> gives a good solution to this
problem that uses -lincom-, but it would also be nice if -streg-
supported the -noconstant- suboption within options -ancillary()- and

As such, we have added this to our list for future development.

*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index