|
PureBytes Links
Trading Reference Links
|
First a little philosophy (ideas rule!).
Words do not have absolute meanings.
They can have different meanings to different people and also vary
depending on the context in which they are used.
The most important thing is to be 100% clear on what they mean to you.
Don't be afraid to step outside the box and define things according
to your own perspective (of course the best place to start is by
following the *experts* and then later on, once we have more
experience,setting out to prove them wrong).
In the context of trading, words have a utility function, so in the
end it all has to make us money; that is the final measure of how
correct our definitions are.
If you are making money, no-one can say you are wrong.
The real difficulty starts when we want to communicate with others.
Then we have to try to make sure we are talking about the same things
and using the same definitions, but more often than not, mis-
understanding is the end result (I guess we must like or benefit from
the social interaction or we wouldn't bother at all).
*******************************************************************
The Chadwick definition of a system is not particularly useful to
traders.
An approximate definition could be that a system is the process we
follow to apply a strategy intended to achieve our goal.
It is better if the goal is simply and clearly defined ( a finite or
small target).
Strategies and processes can be complex and goals can be broken up
into sub goals but that is a longer story.
It is not dissimilar to an algorithm in computing, which is why
programmers are ideally placed to become successful traders (or
should be).
Buy and hold, evaluating stocks on fundamental criteria or throwing a
dart at a stocklist pinned on the wall are all systems.
The interesting question is;*Which system is more successful,
according to our definition of success?*.
>From your intitial approach it appears that your pre-deliction is
towards mechanical system trading.
The strength of MST is that the rules are simple, generally few in
number and 100% objective.
The fun part is in defining the goals and strategies which can be
decidedly complex and/or subtle.
While websites, dictionaries etc are very good reference sources, the
quality is often lacking, especially when it comes to trading
concepts.
I would recommend that texts written by traders are the best place to
start.
Very few of the textbooks rate 10/10 but they do set us on the right
path.
For mechancical traders *Design, Testing and Optimisation of Trading
Systems*, by Robert Pardo is a reasonable start.
Personally I only rate it 6/10 but the average book is somewhere
below that.
I disagree with some of his conclusions but other traders argue
vehemently in support of his approach.
********************************************************************
Specifics:
Am I on the right track?
Definitely.
It's a long way to the top though.
Suggest improvements to your approach?
Keep doing what you are doing (reading, thinking, questioning,
discussing, trialling; you are in good company with AmiBroker).
It will be interesting to see what you can do in a few years if you
are still around.
Suggest improvements to your system development?
I would say to think carefully about the criteria you use to evaluate
your systems (there is discussion in Pardo about this) and to
understand the principles of money management thoroughly (there are
1000's of systems but you only have one trading bank).
Random observations:
1. There are indicators that fall outside the categories you mention
e.g broadmarket indicators (put/call ratios, advances/declines etc)
and economic indicators (interest rate changes etc).
2. Exits can also be on a profit stop, reversal of signal, time stop,
number of bars stop and others.
3. Volatility stops might be more useful than % stops e.g a -5% stop
on a slow moving stock does not equal a -5% stop on a fast moving
stock (fast moving = = volatility?)
4. Volume looks like one of the easiest indicators to exploit but I
have found it to be harder to handle than it looks e.g there is a
buyer for every seller so when prices move up on high volume there
are just as many traders wanting to get out as those wanting to get
in - which group, if any, is correct in their view of future price
movement?
5. It is easy for commentators (websites, authors, forum members) to
make statements like:
*Many times before a stock advances, there will be a period of
increased volume just prior to the move.*, but how true is that?
How can we prove if that statement, or any trading argument, is true
or false?
Could any two traders reach agreement on that point?
6. The MA approach to divergence is interesting.
I don't know if it would lead to a profitable system or not.
Personally I have a bias against MA's because they are always late to
signal a change.
Looking at the second chartshot you provided it is obvious that the
AccDist line is *trending* up or has made a reversal, before the
shorttermMA crosses above the LTMA (for this example I am referring
to the second cross).
The chart also indicates the problem that MA's pose for analysts when
MA crossovers are very short lived (the first cross in your chart is
an example of this).
MA's aren't alone in this area though; indecisive or sideways
movement is a tester for all TA systems.
That doesn't mean that MA systems can't or aren't used by profitable
traders.
7. I wouldn't say that divergence can't be coded in Ami.
I haven't tried but if you search the archives you might find some
code or the forum might provide some if you specifically ask them how
to do that.
8. I find that divergence is often quoted as an entry or exit signal
but I nearly always find that the examples given are very subjective
e.g in the MSFT chart you provide looking at the first long divergenc
example. The price action confirms that the stock is potentially at a
bottom (double bottom in a narrow range) and the yellow line as drawn
indicates this as it is a sideways *trend*.
Immediately after the L line there is a bullish gap *breakout* above
the previous minor peak (a potential upward price move).
The AccDis upward move, as drawn for the same period, doesn't
actually confirm this until the same bar, so it is not exactly news.
Some traders, who use multiple signals, would say that it is a
confirmation signal (fair enough if it pays the bills).
If we go back a few bars to the start of the yellow line, we could
say that the AccDis line went to a minor bull move ahead of the price
action (at the point where the price has a bear spike).
If that was the chosen signal the long entry would have been back
then in a *brave* contrary (buy the low) trade.
Once again, fair enough, but one swallow doesn't make a summer.
If this can be coded and tested and it is a consistent *winner*;
happy days!
The other two examples can be similarly critiqued.
(keep in mind I don't have your chart other than in word, and
consequently I don't have a crosshair cursor to line up the bars so I
am just giving an off the cuff example of critical analysis of
apparent truths).
9. One of my personal biases is to a purist approach where I use data
at the most fundamental level that I can.
I have a distrust of indicators on indicators (once again if it can
be made to pay; great!) e.g price is the primary level of data and
AccDist is the first derivation followed by an MA on the AccDist
which is a second order derivation.
To me this increases lag and what I call *sloppiness* in the signals.
The analogy is if you pick up an apple off the table with your hand
that is a fundamental human movement.
If you build a mechanical arm with two joints and extension pieces
and use that to pick up the apple, sloppiness in the joints and
distance from the apple makes the task more difficult to perform with
accuracy.
By analogy the more complex the indicator the greater degree of lag
and potential error.
That of course is a personal bias and other traders would probably
tick me off about that one.
Good luck with your trading.
I am flat out at the moment so I doubt if I will be able to comment
on your next topic.
Regards,
Brianzee123
--- In amibroker@xxxxxxxxxxxxxxx, Bill Halliday <halliday_mo@xxx>
wrote:
>
> Hello,
>
> I've placed a .doc file in the files area called 'My First
system'.
>
> Please review it and answer the following questions:
>
> 1) am I on the right track?
> 2) can you suggest any improvements to my approach to developing
a system?
> 3) can you suggest any improvements in this system?
>
> Send your answers directly to me at: halliday_mo@xxx .
>
> Thanks,
>
> Bill
>
> PS Tomasz - Please dispose of the coppy I sent you, I made many
revisions to it's presentation in the copy I placed in the files area.
>
>
> ---------------------------------
> Need Mail bonding?
> Go to the Yahoo! Mail Q&A for great tips from Yahoo! Answers users.
>
Content-Description: "AVG certification"
No virus found in this incoming message.
Checked by AVG Free Edition.
Version: 7.5.432 / Virus Database: 268.17.20/664 - Release Date: 2/2/2007 3:42 PM
|