using loglog link in VGAM or creating loglog link for GLM

torbjore

Sun, Jan 23, 2011 10:56 AM #

I think you guys make it more difficult than it has to be. Estimating
probability of success with a loglog-link is equivalent to estimating
probability of failure with a cloglog-link, so all you have to do is to
change the response variable accordingly (and then you can interpret
parameter estimates (or any contrasts in general) as hazard ratios)...

Though, I wish there was a loglog-link in glm and lmer so you wouldn't have
to do this.

View this message in context: http://r.789695.n4.nabble.com/using-loglog-link-in-VGAM-or-creating-loglog-link-for-GLM-tp898386p3232866.html
Sent from the R help mailing list archive at Nabble.com.

Gavin Simpson

Mon, Jan 24, 2011 4:26 AM #

On Sun, 2011-01-23 at 10:56 -0800, torbjore wrote:

Err, isn't that **exactly** what I said?! (May be the "you guys" wasn't
aimed at me but there were only two responses in that thread, one of
them from me... and you haven't quoted to what it was you were
referring.)

You are, in effect, asking other people to maintain more code for you
when you can easily alter the coding for the response.

G

%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%
 Dr. Gavin Simpson             [t] +44 (0)20 7679 0522
 ECRC, UCL Geography,          [f] +44 (0)20 7679 0565
 Pearson Building,             [e] gavin.simpsonATNOSPAMucl.ac.uk
 Gower Street, London          [w] http://www.ucl.ac.uk/~ucfagls/
 UK. WC1E 6BT.                 [w] http://www.freshwaters.org.uk
%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%