|
PureBytes Links
Trading Reference Links
|
Nicholas
This turned into a really long response, so you might want to delete it
after the next paragraph. What follows is a sort of description of how we
started, what we do and how, without going into details of our proprietary
indicators. Anyway, here's my long winded response.
Neither.
We have our own indicators that are basically derived from price movement.
One of our basic assumptions is that whatever's happening in the market will
be reflected in the action of the price.
In addition, we feel that with all of the available computing power out
there, everybody can massage the 'standard' or prepackaged indicators and
experts and develop their own systems. We just don't know how valuable they
are. We have spent the last 50 years developing our own indicators. I have
to admit that the basic components were built in the early 50's by trial and
error. We went through the graphical approach, Point & Figure, manual work
sheets, etc. I remember using pieces of cardboard that were cut out in
order to 'automate' our ability to calculate modified moving averages of
some of our indicators.
I think that if you are going to be a 'technician' you need to go through
this methodology in order to develop your own 'feel' for the market. This
feel will help you to develop your own techniques. We made a quantum leap
with our ability to analyze our indicators and test out our theories when I
took my first job out of college with IBM in 1961. Back then, we were
probably years ahead of most NY or Chicago trading firms who were just
trying to automate their accounting and payroll.
Eventually, everyone blew by us, but by then we had our core indicators and
were operating in a unique environment. Our approach was different than
theirs. They utilized a lot of theoretical statistical and mathematical
calculations, looked at a whole bunch of technical approaches like
arbitrage, etc. We used a more simplistic approach based upon what
developed to operate our manual approach in the 50's.
We do have a pretty good math and statistical background. My dad, who
started us in this business, actually he started some 60 or 70 years ago,
has a Masters in Statistical Economics and has forgotten more about numbers
than I'll ever remember.
He's celebrating his 90th birthday this June and he still develops trading
systems using a Clipper based system (started as a dBase II system) on PCs
in the mid 70's (way before Apple and IBM were even in the market). He
still writes code and does basic testing. When he finds something that
'feels' or looks right, he e-mails it to me. My contribution lies in being
able to 'sense' relationships, so to speak. Again, there is a lot of art to
this, based upon my 45 years experience. My brother then follows up and
does the back testing and detailed analysis. This separation of roles
enables us to really test concepts and new indicators without the built in
bias that a developer has. Occasionally, he will give me some feedback on
something he spots and I go back and review it. For some reason,
relationships seem to jump off of the paper at me and I'm able to
intuitively determine whether or not there is a possibility there or whether
we need to modify it, or whether we need to scrap it. It's not uncommon for
my dad to send me something that he's put together and when I run it, I have
absolutely no idea what in heck he saw in it in the first place. I'll
e-mail him back my comments, like "what are you talking about here?" And "I
don't see anything here, what signals are you looking at?" In actual fact,
he e-mails me programs with no instructions or comments. Sometimes he'll
program in some B or S signals or overbought or oversold indicators, but I'm
usually on my own. I run the programs, looking at various commodities,
trying to determine whether anything there makes sense or whether he's
getting even more senile <G>. Then I'll call him and ask what his thoughts
were when he wrote this. The response is usually, "Oh, forget that one,
I've got something better I'll send you" or in rare instances, he might have
flagged something interesting. In other instances, I ignore his 'signals'
and see something totally different in the numbers.
Our current system started out as one of his 'ideas' which we then messaged
until it felt right. It worked great in Clipper, so I took the time to try
to program it into Metastock. Couldn't do it at that time due to the lack
of computational capabilities of Metastock. Forced me to take a couple of
months to develop an alternative calculation of our momentum indicators (we
have 2 each representing up and down momentum. What I developed looked
pretty good and when combined with the original indicators, looked even
better. I then went back and programmed the original Clipper indicator into
TAS. More recently, I've programmed it into Excel and am looking at Visual
Basic. I have saved all of Walter's (and everybody's Excel posts) and am
getting ready to try to figure out how to use Visual Basic and/or Visual
Basic for Applications.
After a few years, we realized that Bonds appeared to be related to our
trading the S&P futures and kept us out of trouble and eliminated some bad
trades. I did identify, and posted here for discussion last year, the
possibility of the link between equities and bond disappearing. And the
relinking of Equities and Bonds early this year. This has only happened
twice in the past 17 years, as far as I've can see.
You'll need to constantly monitor your trading activity and tweak your
systems to meet the current market environment.
Anyway, I guess what I'm saying is that you need a 'feel' for what you do
and you need to develop indicators or use packaged indicators that you are
comfortable with.
With ours, we're not in conflict with anybody else. Very rarely do out
trades coincide with other people, whether that's good or bad I'm not sure.
I apologize for the long response, but it's late, I'm tired and ready for
bed. My 8 year old and I went to the Cardinals/Dodger's game tonight as
well as dinner at the Stadium Club celebrating his buddy's ninth birthday.
We did see a major league record made. One of the Cardinals who's name I
can't remember - (Bettig?) hit 2 grand slam home runs in the same inning
(and off the same pitcher). This has never been done before and the fans
gave him a standing ovation.
Anyway, I don't know what else to say other than I'm not sure whether
anybody can develop technical indicators, that are fixed in concrete, and
will work on all commodities. We have felt for a long time that there is a
certain amount of art in this 'science.'
A final note. I traded for 2 years on paper, until I was 15 or 16, before
my dad let me make my first 'real' trade. I watched the market, played with
the numbers, made a lot of mistakes that didn't cost me any money. Can't
say enough about taking your time before entering the market. Remember we
are in an unprecedented bull market, which one of these days will turn. You
need to be able to trade in a more real world market.
Regards
Guy
-----Original Message-----
From: owner-metastock@xxxxxxxxxxxxx
[mailto:owner-metastock@xxxxxxxxxxxxx]On Behalf Of Nicholas Kormanik
Sent: Friday, April 23, 1999 4:20 PM
To: metastock@xxxxxxxxxxxxx
Subject: RE: Some indicators should work....
"...this [system adjustment] was basically changing how we **weighted** our
various indicators."
Guy,
What one or two indicators do you give the most weight to? MACD?
Stochastics?
Thanks,
Nicholas
|