[R] non-linear regression and root finding

[Berwin A Turlach]

G'day Troels,

On Tue, 7 Nov 2023 07:14:02 +0100
Troels Ring <tring using gvdnet.dk> wrote:

> Be as it may, I wonder if not your method might work if only we KNOW
> that pK1 is either positive OR negative, in which case we have pK1 =
> -exp(theta1)?

If pK1 can be either negative or positive (or 0 :-) ), and it is just
the ordering that you want to have enforced, then I would try the
parameterisation:

pK1 = pK1
pK2 = pK1 + exp(theta2)
pK3 = pk2 + exp(theta3)

and optimise over pK1, theta2 and theta3.  

As long as you want to know the estimates only.  

Asking for standard errors of the original estimates would open another
can of worms. :-)

Cheers,

	Berwin