R help - Jul 2004 - solving for a transition point in a piecewise nonlinear model

I want to fit a piecewise nonlinear model in which the transition point is 
an estimated parameter.

Something like:

F(x)=F1(x) when x>alpha,
F(x)=F2(x) when x<=alpha.

How can I solve for alpha within the nls call?

Thanks,
Rob

Dear Robert,
In general it may be difficult to estimate a model with generic (possibly
nonlinear) functions before/after the changepoint  to be estimated too.

However if you are willing to make some restrinctions on your F1(.) and
F2(.), you could semplify the problem..

For instance, have a look to the strucchange and segmented packages. The
former mainly deals with time series data, the latter with GLM.

hope this helps,
vito

PS Of course you can always perform a grid-search algorithm by fitting
different models for each fixed alpha, and picking the value that maximizes
the Lik..



----- Original Message -----
From: Robert Musk <rmusk at postoffice.utas.edu.au>
To: <r-help at stat.math.ethz.ch>
Sent: Wednesday, July 14, 2004 6:43 AM
Subject: [R] solving for a transition point in a piecewise nonlinear model

> I want to fit a piecewise nonlinear model in which the transition point is
> an estimated parameter.
>
> Something like:
>
> F(x)=F1(x) when x>alpha,
> F(x)=F2(x) when x<=alpha.
>
> How can I solve for alpha within the nls call?
>
> Thanks,
> Rob
>
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://www.stat.math.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html