R help - Jun 2004 - interval regression

Hi,
 
does anyone have a quick answer to the question of how to carry out
interval regression in R. I have found "ordered logit" and
"ordered
probit" as well as multinomial logit etc. The thing is, though, that I
want to apply logit/probit to interval-coded data and I know the cell
limits which are used to turn the quantitative response into an ordered
factor. Hence, it does not make sense to estimate the cell limits (e.g.
in zeta for the function polr).
 
Thanks,
 
Thomas Ribarits
 

	[[alternative HTML version deleted]]

I believe from your description that this is interval-censored survival
and can be handled by survreg.

It would also be very easy to amend polr to do this.

On Wed, 30 Jun 2004, Thomas Ribarits wrote:
> does anyone have a quick answer to the question of how to carry out
> interval regression in R. I have found "ordered logit" and
"ordered
> probit" as well as multinomial logit etc. The thing is, though, that I
> want to apply logit/probit to interval-coded data and I know the cell
> limits which are used to turn the quantitative response into an ordered
> factor. Hence, it does not make sense to estimate the cell limits (e.g.
> in zeta for the function polr).
-- 
Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272866 (PA)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595