stableFit

[Yohan Chalabi]

>>>> "RL" == Remigijus Lapinskas <remigijus.lapinskas at mif.vu.lt>
>>>> on Tue, 06 May 2008 08:54:13 +0300


 RL> Dear Maintainers of Rmetrics,
 RL> 
 RL> I use Windows XP, R 2.6.1 and fBasics 270.73
 RL> 
 RL> I want to estimate the stability parameter of a stable
 RL> distribution with
 RL> stableFit. As you can see, I receive three different answers
 RL> for three
 RL> calls to stableFit. Is it a bug or a feature?


Dear Remigijus,

If you replace in function '.qStableFit' all calls to 'approx' with
'spline', it does improve the estimation.

Note that 1000 points might not be enough to estimate the parameters.
I would recommend you to use more points.

I will see if we change this in the next release.

sorry for the late response,
Yohan

-- 
PhD student
Swiss Federal Institute of Technology
Zurich

www.ethz.ch
www.rmetrics.org

NOTE:
Rmetrics Workshop: http://www.rmetrics.org/meielisalp.htm
June 29th - July 3rd Meielisalp, Lake Thune, Switzerland