|
PureBytes Links
Trading Reference Links
|
|
That's a keeper!!
----- Original Message -----
Sent: Friday, February 04, 2005 11:31
AM
Subject: [Metastockusers] Re: Time to
ditch MetaStock?
Kut2k2, it seems that you've missed the gist of my
post.
Your intense diatribe seems typical of someone that is well
entrenched
in his/her attachment to a particular idea or piece of
software.
> Face it, Jose, you have more than an emotional
investment, you have
> a financial investment in MS as
well.
Unlike yourself, I don't have an emotional investment in any
software
all - I just happen to use what works best for me at the
time.
Linux is far superior to Windows, yet the latter is more useful and
productive for me at the moment. The same applies to any AB vs MS
argument - you will find none from me.
I'm not interested in either
being a software groupie, nor defending a
piece of code to the
death.
My investment is on life. Time is shorter and much more
valuable than
most people believe.
When it comes to trading, I just
can't find the time nor the patience
to dedicate my life and energy to
purely intellectual pursuits.
I certainly don't have the freely available
time & patience to turn
charting into a geeky hobby.
My
"financial investment" in MetaStock happens to be a side-effect of
trading. I enjoy helping others that are prepared to help
themselves,
and will continue to do so for as long as I
can.
Furthermore, I feel that I have a responsibility to not only help
others, but to point out the traps and mistakes which I have already
been through myself.
> Basic building blocks don't mean
squat if you can't apply them in
> innovative ways.
Innovation
comes from within, not from charting software.
> That, plus all
the nightmares I keep hearing about the system tester
> make me glad I
never upgraded my MS and pushes me more and more
> towards
AB.
The system tester is the most overrated and useless tool in both MS
&
AB, for different reasons. Optimizing/curve-fitting system
parameters
& backtesting without realistic expectations, no matter how
impressively complex it may seem, are the biggest enemies of a true
system developer.
If you want real backtesting without associated
distractions, try
TradeSim.
> One security at a time.
Not real appealing for people who want to
> trade a portfolio. Think
about it.
Trading a portfolio and developing strategies that involve
other
correlated instruments, is simply not the same process. Think
about
it. ;)
> Ever want to try fixed-fractional
position sizing your trades? Good
> luck doing that with
MS.
I can do this without any problem in MS.
> Did Equis
ever correct the code for vhf()? Just morbidly curious.
Just
morbidly curious - would you happen to know what to do with the
Vertical
Horizontal Filter? If so, how would you use it in your
trading
strategy? Ask yourself: could this perhaps be another
intellectual
pursuit that has little or no relevance to real
trading?
>...given that I KNOW the BHI is garbage,
My
advice here is: don't get caught-up on jargon and processes that
have little or no relevance to real-world trading. Use the tools
that
you are most comfortable with, which can extend and realize your
potential without bogging you down with unproductive
pursuits.
> If you can't implement those thoughts because of a
limited MFL, what
> good is it?
You see, a limited formula
language can have some unexpected
usefulness - it can help prevent a
potential trader from turning the
whole system development process into a
complex intellectual pastime
that has little or no relevance to the real
trading world. One only
has to look at the thousands of useless
indicators available, to
understand this.
Let's face it: if you are
really interested in developing a successful
trading strategy, you
certainly don't want to spend a great proportion
of your life with coding
bits and pieces that have little or no
relevance to the real
markets. A degree in rocket science is not a
prerequisite
here.
> For those of us who don't, some of your points don't
make
> a lot of sense.
Take a deep breath, a step or two back,
and you may catch a glimpse of
the bigger picture.
Attachment to
any specific idea, guru, or piece of software, is deadly
in this
game. A warning sign is a lack of humor, and taking offense
(even if
none was intended) when one's favourite software seems to be
attacked. Let go, and you'll become both a better trader and human
being.
All the power of modern hardware & software, is not
really necessary
to become a good trader: by all means, look for
help out there, but
search for the answers within.
jose
'-)
http://www.metastocktools.com
---
In Metastockusers@xxxxxxxxxxxxxxx, "kut2k2" <kut2k2@xxxx> wrote:
>
> --- In Metastockusers@xxxxxxxxxxxxxxx, "Jose"
<josesilva22@xxxx>
> wrote:
>
> There seems to be a
lot of recent discontent building up with
> regards to the lack of
MetaStock development from Equis/Reuters.
> It is indeed true that other
charting programs have overtaken MS in
> a major way.
>
>
For example, AmiBroker (AB) can plot a composite system equity curve
>
based on hundreds of securities, and allows the changing of
> parameters
with the aid of real-time variable sliders.
> Watching the AB composite
equity plot change in real-time is a
> marvel.
> Just getting
MetaStock to plot the same single composite equity
> (forget real-time
manipulation), involves so much processing
> overheads that it is not
practical.
>
> So, AmiBroker is indeed superior to MetaStock in
many ways.
> However... I still find myself using MetaStock for 95% of
my tasks.
>
> Why?
>
> 1) Familiarity &
ease of use = time-saving - it generally takes me
> five times longer
to code in AB;
>
> 2) MetaStock's Graphic User Interface is still
superior to AB;
>
> 3) More than 95% of programming tasks are
basic building blocks
> which don't require complex
programming;
>
> Basic building blocks don't mean squat if you
can't apply them in
> innovative ways. You certainly won't get
far anymore applying them
> in conventional ways, even if at one time
you could. MetaStock's
> shameful refusal to deal with their fixed
parameter problem is a
> case in point.
> AB has allowed variable
lookback periods in all their lookback
> functions for years now.
The LastValue(A+PREV-PREV) fix in MS is a
> slow-motion bad joke, and
it doesn't even work for a lot for
> lookback functions that I need it
for. This alone makes MS last
> millenium's technology.
That, plus all the nightmares I keep
> hearing about the system tester
make me glad I never upgraded my MS
> and pushes me more and more
towards AB.
>
> That, plus the fact that some of those basic
building blocks in MS
> are erroneous. Did Equis ever correct the
code for vhf()? Just
> morbidly curious.
>
> Does the
Buy/Hold Index in the Reports of the latest MS version
> still give a
garbage result? Again, just morbidly curious. I keep
> wondering
every time I run a test how valid my results are, given
> that I KNOW
the BHI is garbage, and I worry what else is garbage
> output.
>
>
> 4) Successful mechanical trading strategies are composed of
60~80%
> logical thought, and 20~40% actual programming skills.
>
> If you can't implement those thoughts because of a limited MFL, what
> good is it?
>
> Ever want to try fixed-fractional
position sizing your trades? Good
> luck doing that with
MS.
>
>
> It is the last point that needs to be generally
acknowledged &
> understood.
>
> It is not the actual
charting package that decides how successful
> the development of one's
mechanical or subjective trading system may
> be, but rather, it's the
trader's own ideas and ability to put them
> into a logical &
codeable form that makes a good system worth
> considering.
> More
programming complexity is not always better.
>
> Ditto for less
programming capability.
>
>
> If I were to be just
interested in the intellectual challenge of
> coding difficult or
abstract ideas, I would have probably left this
> group to join the
AmiBroker camp a long time ago, just as others
> have already
done.
>
> However, being much more interested in the bigger
picture, means
> that the charting tool is not so important to
me. MetaStock is fine
> for just about all system development
purposes.
>
> One security at a time. Not real appealing
for people who want to
> trade a portfolio. Think about it.
>
>
> If productivity is also important to you, and you already
use
> MetaStock, you may find that there is little or no need for
better/
> faster tools, or for re-learning a new process.
> If you
are not using a charting package yet, consider MetaStock's
> easy
learning curve before choosing a career in rocket science.
>
>
Take care not to lose sight of the forest for the trees. :)
>
>
> jose '-)
> http://www.metastocktools.com
>
> Face it, Jose, you have more than an emotional investment, you
have
> a financial investment in MS as well. For those of us who
don't,
> some of your points don't make a lot of
sense.
Yahoo! Groups Links
|
|