On 6 May 98 at 10:59, Petr Kuzmic wrote:
> and most people -still- use 'formal standard errors' of fitting
> parameters, instead of full confidence intervals! This is actually
> quite absurd, but I'd venture to say it'll go on forever.
Petr, you are dreaming about a perfect world... I would even be happy if
most people used nonlinear regression, rather than Linewaever-Burk etc!
But on the topic of confidence intervals, as far as I can see these are
only easy to obtain if one has used a method like Levenberg-Marquadt. This
produces, as a "secondary" product, estimates of the co-variances. Many
other methods of minimisation do not, and in that case one has to obtain
confidence intervals by Monte Carlo simulation. The problem is that the
Monte Carlo simulations spend a lot more computational time than was
needed to obtain the parameter estimates in the first place.
Pedro
Pedro Mendes Inst. Biological Sciences
prm at aber.ac.uk (home: pedro at enzyme.demon.co.uk) Univ. Wales, Aberystwyth,
http://gepasi.dbs.aber.ac.uk/pedro/prmhome.htm Ceredigion SY23 3DD, U.K.
http://gepasi.dbs.aber.ac.uk/softw/gepasi.html FAX: +44 (0)1970 622307
