.. ok, needed a bit and a second look. I was puzzled by the quality of
the approach. It was too good to be true.
While I did not get exactly what they are doing, they are clearly
looking at all the data. So they are NOT doing a
time series prediction. Any time series prediction approach cannot go
so sharply from -1 to 1.
If it would, it would also lead to a significant amplfication of any
noise.

So, yes. This is a very interesting approach and paper. Perhaps it can
be adapted and extended into a time-series
prediction. But I do not see a direct relation to the question.

Klaus


On May 21, 8:43 am, Klaus <[email protected]> wrote:
> Dear Nonlinear,
>
> this paper is indeed impressive. He also studies explicitly the
> question how to approximate the
> first derivative. The figures and result are extremely good..
> - what I did not get was the relation to double exponential
> regression.
> Is it hidden in some of the referenced approaches (I do not know?)
> My impression was that the approach presented would be multiple orders
> of magnitude more complex.
> (though given the quality of the results, still a valid goal..)
>
> Cheers
>  Klaus
>
> On May 11, 11:43 pm, nonlinear <[email protected]> wrote:
>
>
>
>
>
>
>
> > On Wednesday, May 11, 2011 5:33:55 PM UTC-4, Alexana wrote:
>
> > >   If you are looking for a bit of mathematical theory on this type of
> > > methodology, take a look at the attached pdf.
> > >  **
>
> > ... and 
> > this:http://math.lanl.gov/Research/Publications/Docs/chartrand-2010-numeri...

-- 
You received this message because you are subscribed to the Google Groups 
"JBookTrader" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/jbooktrader?hl=en.

Reply via email to