Re: Fwd: [ferret_users] fourth-order polynomial least square

[Peter Szabo <szabpet83@xxxxxxxxx>]

Dear Patrick,

thanks a lot for your detailed help.

I could make it work, but i would like to fit my time-series using ordinary least squares with a forth-order polynomial over many (like 100) years- which is more like a trend line, I guess.
I assume that changing any of the arguments of the 4 different methods would not give me what I wanted, or am I wrong?

Thanks, 

Peter

On Thu, Mar 17, 2011 at 11:51 AM, Brockmann Patrick <Patrick.Brockmann@xxxxxx> wrote:

--