Re: [Quantlib-dev] GARCH by VB.NET

[Joseph W. <jo...@gn...>]

The idea was that you'd put in a time series into calibrate which would set=
=20
the parameters for the GARCH model.  You could then use the object for a=20
different set of date.


=D4=DA Tuesday 18 July 2006 20:55=A3=AC=C4=FA=D0=B4=B5=C0=A3=BA
> Hi Joseph,
>
> One question:
> What does the function - calibrate() work for?
> Is it for the "maximum likelihood estimation" of parameters?
> thanks a lot.
>
>
> Regards,
>
> Charles
>
> On 7/18/06, Joseph Wang <jo...@gn...> wrote:
> > Charles Wings was in the process of implementing GARCH(1,1) in C++ using
> > code that is half written there.  You might want to get in touch with
> > him to see if you two can cooperate.
> >
> > =D4=DA 2006-07-17=D2=BB=B5=C4 14:45 +0100=A3=ACPatrick X. Zhang=D0=B4=
=B5=C0=A3=BA
> >
> > > Hello Joe:
> > >
> > > Read your thread on Wilmott about "Weirdness in Shanghai warrants":
> > >
> > >
> > >                      Fri Apr 21, 06
> > >                           04:05 PM
> > >                     User is offline
> > >                          View users
> > >                            profile
> > >
> > >                     I'm starting to
> > >                    finally get some
> > >                          results in
> > >                          looking at
> > >                            Shanghai
> > >                       warrants, and
> > >                       they are very
> > >                               odd.
> > >
> > >                     I've calculated
> > >                    the implied vols
> > >                   of a put warrant,
> > >                     a call warrant,
> > >                             and the
> > >                    underlying using
> > >                     constant weight
> > >                   (I'll try to do a
> > >                   GARCH calculation
> > >                    next week once I
> > >                     code up a GARCH
> > >                            class in
> > >                          QuantLib).
> > >
> > >                              ......
> > >
> > >
> > >                   Another question.
> > >
> > >                          Has anyone
> > >                         written and
> > >                       freeware plot
> > >                      templates in R
> > >                           that plot
> > >                         candlestick
> > >                     diagrams and/or
> > > variograms/coorelationgrams? If not, I'll delve into the guts of R and
> > > write some.
> > >
> > >
> > >
> > >
> > >
> > > Just wanted to know whether you have done the GARCH part in C++ or R -
> > > I'm trying to implement GARCH(1,1) in VB.NET and obviously the problem
> > > is the maximization/minimization routine (if we don't use Excel
> > > Solver).  Any thoughts?  Someone else said R could do it but I have no
> > > idea how ;-(  Thanks a lot.
> > >
> > > Best,
> > > Pat