[R] nls: singular convergeance

[Rolf Turner]

On 13/10/11 18:02, Redding, Matthew wrote:
> Dear R-experts,
>
> I have 28 data points that I would like to fit with a non linear
> broken-stick -- with three fitted parameters.
> When I view trace -- and use the final values as lines on the graph of
> data -- it looks pretty good.
>
> Q1. Why am I getting singular convergeance?
> Q2. Can you suggest another approach that may prove more satisfying?
<SNIP>

Have you thought of using the "segmented" package instead
of rolling your own with nls?

     cheers,

         Rolf Turner