|
PureBytes Links
Trading Reference Links
|
Brian:
I have no problem with advising people to look before
they leap so I guess I misinterpreted your comments as pontification based
on assumption/opinion.
As for your last comments about Gerard's code I wonder
why you say that it is not a system. I see in the code up and down
arrows. Sounds like a system to me although it might not be up to your
standards.
Your comments about not being efficient (whatever that
is) or simple (whatever that is) or not being optimized (Gerard does note use of
manual optimization and the need to adjust parameters for different markets) are
irrelevant, imo. What counts is bottom line results which I gather you
have not evaluated either by itself or relative to other
systems/indicators/methodologies. In my world, downgrading based on
subjective arguments are taken with a grain of salt.
There are by some counts seven categories of
indicators, including momentum. Within each category there is much
similarity in results and interpretation. I have not seen any objective
basis for throwing out, for example, CCI in favor of, for example, price
action. As I have mentioned on several occasions, imo, the way to go is to
combine charts (patterns, trendlines, S&R, cycles, etc.) with a variety of
indicators, requiring a certain degree of convergence - but then I'm a
discretionary trader so this would not work for many. Of course, no
system/methodology/philosophy is the "word" and we all take what is useful and
leave the rest alone. From what I have seen most
systems/methodologies when properly executed produce satisfactory results and
that's all that counts in my book.
Good trading.
Bill
----- Original Message -----
Sent: Wednesday, December 19, 2007 5:02
AM
Subject: [amibroker] Re: Dow Theory
AFL
>
Bill,
>
> Anyone using the signal, or thinking of using it might
benefit from
> looking a bit further (that is all I am saying).
>
> BTW it is not a system as it doesn't have exits etc.
>
>
What I am pointing out is:
>
> 1) The 'trade', as coded by Gerard,
is a close cousin of the
> InverseRTM so additional commentary can be
found under that heading.
>
> 2) The exact signals, as produced by
Gerards code, can be obtained by
> other simpler means.
>
>
3) Alternative ways of getting that signal are more efficient.
>
>
IMO it is not optimized as written at present.
>
> Simplifying the
code and cutting it lose from the CCI positions it
> correctly within the
framework of its trading philosophy and that
> then leads to ideas that
help towards optimizing the trade.
> Getting rid of the CCI also fits
nicely with my views that price
> action is a superior 'indicator' to
many of the price derived
> indicators (CCI is highly over-rated by
some?).
>
> Since my friend Bob is counting my 'kernels of wheat' I
made sure I
> put one in there for him (I dehusked if first
though).
>
> brian_z
>
> P.S
>
> Since
it is philosphy that distinguishes Mankind from the animals:
>
>
"One mans chaff is another mans mattress".
>
>
>
>
> --- In amibroker@xxxxxxxxxxxxxxx,
"fimdot" <fimdot@xxx> wrote:
>>
>> Assumptions and
academic exercises aside, all that matters is
> whether
>> the
system/methodology produces satisfactory results for the user.
>> That's the bottom line and the rest takes second place.
>>
>> Bill
>>
>> --- In amibroker@xxxxxxxxxxxxxxx,
"brian_z111" <brian_z111@> wrote:
>> >
>> >
>Look in Files Section - Listing Dow Waves by Gerard Carey
>>
excellent.
>> >
>> > No disrespect to Gerard Carey but
I don't rate this as an
> excellent
>> > TA study. It is the
start of an interesting discussion and I do
>> thank
>> >
Gerard for sharing it.
>> >
>> > As I pointed out
to Gerard at the time, the Dow Indicator entry
>> > signals are
very similar to an Inverse Reversion To Mean System,
>> > which is
bread and butter in the industry.
>> > They always look great on a
chart with a longer than average
> trend
>> in
>>
> place (which obviously occurs less than 50% of the time).
>> >
>> > From memory, if you look at Gerard's example chart of the Dow
>> > Indicator plotted on the ^DJI it makes 12 - 13 successive
touches
>> of
>> > a bullish MA.
>> >
>> > My propostion regarding waves, or cycles, is that if you
measure
>> > their frequency, and magnitude, you can expect to get
a lot of
>> > variance and something approaching a bell curve if
they are
>> > distributed (either frequency or
magnitude).
>> >
>> > So 12 - 13 successive touches of
a bullish MA that continue on to
> a
>> > higher high is not
something that happens all that frequently.
>> > To provide some
balance look at the same indicator on some other
>> > charts, or
better still backtest it to obtain a larger dataset.
>> >
>> > If we could predict when a stock, or any other instrument, is
> going
>> > into a long term trend (cf a short term trend)
we wouldn't need
> any
>> > other indicators at
all.
>> >
>> >
>> > brian_z
>>
>
>> >
>> >
>> >
>> >
>> >
>> > --- In amibroker@xxxxxxxxxxxxxxx,
"areehoi" <areehoi@> wrote:
>> > >
>> > >
>> > > Look in Files Section - Listing Dow Waves by Gerard Carey
>> > excellent.
>> > >
>> > >
Dick H.
>> > >
>> > > --- In amibroker@xxxxxxxxxxxxxxx,
"greenhorn1983"
> <greenhorn1983@>
>> > >
wrote:
>> > > >
>> > > > --- In
amibroker@xxxxxxxxxxxxxxx,
"MarkK" <MailYahoo@> wrote:
>> > > > >
>>
> > > > Was curious if anyone had written an ALF for Dow theory
>> trades
>> > as of
>> > > >
yet?
>> > > > >
>> > > > >
>> > > > > MarkK
>> > > >
>
>> > > >
>> > > > i've also been
thinking about it lately :)
__._,_.___
Please note that this group is for discussion between users only.
To get support from AmiBroker please send an e-mail directly to
SUPPORT {at} amibroker.com
For NEW RELEASE ANNOUNCEMENTS and other news always check DEVLOG:
http://www.amibroker.com/devlog/
For other support material please check also:
http://www.amibroker.com/support.html
![]()
__,_._,___
|