[scilab-Users] Re: FW: turning point estimation

[Eric Dubois]

Dear Agankideska.

You have to enter peaks and troughs of the reference series and the one you
want to test if it is leading as string vectors of dates.

For instance :
--> Pr=['1992m2';'1994m12';'1998m3';'2000m11';'2002m5';'2004m9'] // peaks
of the reference series
--> Tr=['1991m2';'1993m3';'1996m1';'1999m3';'2001m11';'2003m6'] // troughs
of the reference series
--> Pl=['1994m12';'1997m10';'2000m6';'2002m5';'2004m7'] // peaks of the
leading series
--> Tl=['1993m2';'1996m1';'1998m12';'2001m10';'2003m6'] // troughs of the
leading series
--> banerji(Pr,Tr,Pl,Tl,'lead=4');

(in the text you tried, rfr and rbe are implicitely results tlists coming
from an application of the Bry-Boschan procedure for the determination of
peaks and troughs -function brybos in Grocer-: if you did not run the
function with the data clim_fr and clim_be, it is normal that you cannot
perform the test this way)

Hope it helps

Éric.


2012/3/27 agankideska <agankideska at hotmail.com>

> Dear Eric,
> first i hope, you accept a belated thanks. your help is very important for
> me, thanks a lot more. and also one more question for you abaout benarji
> test. how can we write peaks or throughs date in "rba
> =banerji(rfr('P'),rfr('T'),rbe('P'), rbe('T'),'lead=4');" i tried peaks or
> through date but anyway cant obtain parameter. thanks for your interesting.
>
> --
> View this message in context:
> http://mailinglists.scilab.org/FW-turning-point-estimation-tp3752888p3861916.html
> Sent from the Scilab users - Mailing Lists Archives mailing list archive
> at Nabble.com.
>
> --
> To unsubscribe from this mailing-list, please send an empty mail to
> users-unsubscribe at lists.scilab.org
> To check the archives of this mailing list, see
> http://mailinglists.scilab.org/
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20120327/77697413/attachment.htm>