[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: Re: my statalist posting about time-varying variables in Cox regression

From   Maarten buis <>
To   stata list <>
Subject   st: Re: my statalist posting about time-varying variables in Cox regression
Date   Tue, 10 Jun 2008 17:09:24 +0100 (BST)

--- Allan Garland wrote:
> Thanks for your thoughts on my problem.  I'm not certain about the
> reason for your concerns about what I proposed to do --- the 4
> treatments are completely different therapies, applied in the
> differing epochs.  So, compiling them into a single treatment
> variable seems wrong.  In addition, I think I need 4 separate
> variables because I need to evaluate their separate influences on
> mortality.  
> Therefore, I believe that your second reply seems more appropriate
> --- as long as it is a technically correct implementation of what I'm
> trying to do.  Do you think it is?

I think it is techniqually correct, though you might want to think
about how to interpret differences in effects of these therapies. If
you find differences then these differences could be due to the fact
that the therapies are different or due to the fact that they are
administered at different times. Given your datastructure there isn't
much you can do to disentangle these though...

-- Maarten

Ps. I also sent this answer to statalist as is discussed in section 5
of the statalist FAQ:

Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands

visiting address:
Buitenveldertselaan 3 (Metropolitan), room Z434

+31 20 5986715

Sent from Yahoo! Mail.
A Smarter Email
*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index