|
I started writing my
second article about support and resistance, intending to talk about
cognitive biases, self fulfilling prophecies and the like. Then,
while writing a fairly simple Excel simulation to illustrate a point
I was making, I came across something quite odd. Read on to find out
more.
The simulation was
simple, a completely random ‘market,’ in which at the end of each
interval, the price moves up or down by a random amount. The point I
was hoping to illustrate was that given a
random collection of data that we do not know is random, we will tend
to find patterns in it.
Almost every time I
re-ran my simulation, I was presented with a chart which, had I been
given it and told it was a FOREX pair, I would confidently have drawn
in the support and resistance areas and, given sufficient confluence
with other factors, made trading decision based on. Below are a
couple of the charts generated.
Eventually I'll publish my original
spreadsheet but in the meantime I'd be very interested to see if
anyone reading this has similar or different results when writing
such a simulation; please email any interesting results to
This e-mail address is being protected from spam bots, you need JavaScript enabled to view it
and I'll try to publish them here (giving you credit of course).
Each chart is made up
of thirty two thousand data points, each point can represent the
price based on any unit you may choose; hours, minutes, days...it
makes no difference. Below are the same two charts marked up as I
would the charts for a market I was planning on trading.
On the above charts the
blue lines are where I would have marked an approximate support and
resistance levels and the red lines are ascending trend lines.
This got me thinking
about the reasons I was about to give for support and resistance
occurring and whether, if the same patterns occurred in random data,
need any further explanation need exist?
At this point I should
admit that I’m biased, I want the patterns to be there for a
reason, because this is what makes them predictable and that is what
makes it possible to consistently make money from the markets.
This was particularly
interesting because people are always telling me that trading, in
particularly foreign exchange, is nothing but gambling. If the
patterns I was ‘betting’ on were really nothing but the
by-products of a random process, then these people were right.
On the one hand, the
markets are quite similar to the pseudo-random data generated by a
computer. When a computer generates a random number, it does so using
various equations to generate a sequence of numbers that appear
random to us (like someone reading of a list of number previously
obtained by rolling a dice). Whilst the numbers appear random, they
are actually pre-determined by the underlying process and if all of
the information about the underlying system was available, the next
number in the sequence could be predicted.
A market is similar,
there is an underlying process at work, the buying and selling of
market participants moves the price up and down, if we knew the way
in which these people would behave then we could predict where the
market would move to. Unfortunately the amount of information and the
accuracy required is so great that we cannot accurately predict what
will happen.
On the other hand it is
inaccurate to move from the statement ‘I am observing the same
output from two processes’ to the proposition ‘these two
processes are the same or very similar because they give the same
output.’ This would be similar to saying ‘the sun is yellow, corn
is yellow, and therefore corn is a burning ball of gas.’
That said, it may be
fair to conclude that the two processes have
some element in common, corn isn’t yellow because it’s a ball of
burning gas, but both the corn and sun are yellow because of the way
in which light hitting them is reflected. At this point the
philosophical part of me is tempted to turn this into an article
about consciousness and the Turing Test, this is however meant to be
an article about trading so I’ll save that for a rainy day and
stick to things I think might be useful when trading.
From a practical point
of view, I have no idea what the commone element
is between the markets and my randomly generated markets so, I
compromise. I accept that there probably is some element in common,
but I don’t know what it is, so I make decisions based on what I’ve
observed to work in the past, based on explanations I think work, and
keep in the back of my mind at all times that this element exists,
just to stop myself being too sure I’m right.
This articles fairly
inconclusive because if there is one then I don't know what it is. I wrote it about two months before I
published it and then stalled because I couldn't think of one, if
anyone else can it'd be great to hear it.
For a more general look at the problems with making predictions within financial markets please Click Here.
Please add any comments
below and email any response articles to
This e-mail address is being protected from spam bots, you need JavaScript enabled to view it
and I'll do my best to get them published on here under your name.
|
