[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: [GT] First MAMA-Implementation



On Thu, 2008-03-06 at 23:36 -0600, Weigert, Thomas wrote:
> kw,
> 
> I think it is not wise to not provide the calculate method, as there are
> situations when the indicator might be invoked via this method, e.g.,
> when running anashell.pl (see Analyzers::Process::calc).
> 
> You should, at least, provide a dummy method that calls
> calculate_interval.
> 
> Th.
Hello Thomas,

yes I know and thank you for your advice it motivates me to finish what
I have already started. At the moment the MAMA is rather pre-alpha than
alpha :) I was posting it just to give you an impression of what it
might be. I was never claiming that it is a ready-to-use indicator.
Right now I'm rewriting it to make use of the I:G:Container module.
With Mr. Ehlers methods one can identify a cycle period in the market,
so my plan is to write a MAMA (now maybe following your cookbook),
which can be used on its own and whose sub-indicator period can be
passed to other MAs to make them adaptive this way.

regards,

Karsten

> 
> > -----Original Message-----
> > From: kw [mailto:gt
AT
netsrak.net]
> > Sent: Tuesday, March 04, 2008 9:27 AM
> > To: devel
AT
geniustrader.org
> > Subject: Re: [GT] First MAMA-Implementation
> > 
> > opps ... found a silly mistake in the Hilbert-transform and added a
> > direct start now ... additionally I wrote to Mr. Ehlers to get his
> > recommendations for initialization.
> > 
> > 
> > On Tue, 2008-03-04 at 12:29 +0100, kw wrote:
> > > Following Ehlers Book "Rocket Science for Traders" I implemented a
> first
> > > version of his MAMA ... again the problem of proper indicator
> > > initialization. Right now I would call its status "alpha".
> > > I wrote it without knowing the I:G::Container module.
> > > Thats why there is no calculate routine.
> > > I post it anyways cause it is working and it's way of programming
> might
> > > be an option for quick-testing EasyLanguage coded TradeStation
> > > indicators which can almost be copied line by line due to the
> fifo-lists
> > > I employed.
> > >
> > > regards,
> > >
> > > K.