Seite 62 - E_2012_01

Basic HTML-Version

62
01/2012
www.tradersonline-mag.com
TRADERS´
BASICS
xxxxxxxx
xxxxxxxxxxxxxxxxxxx
Dirk Vandycke
Dirk Vandycke has been
actively and independently
studying the markets since
1995 with a focus on techni-
cal analysis, market dynamics
and behavioural finance. He
writes articles on a regular
basis and develops software,
some of which are available
on his website www.chartmill.
com. Holding master degrees
in both Electronics Enginee-
ring and Computer Science,
he teaches software deve-
lopment and statistics at a
Belgian University. He can be
reached at dirk@monest.net.
This sets stage for our MVI
as an add-on to existing trading
systems or existing indicators,
since no indicator is a complete
trading system in its own right. In
fact, the minute most indicators
make it to the back tests, results
seem to get disappointing very
often and very fast. For the
record, no set of indicators
is a complete system either.
Successful trading needs careful
risk management and consistent
money management discipline.
However, in this article we are
going to look at the effect of
the MVI when added to other
indicators or entry systems. We
look at the Monest Value indicator
as a catalyst.
Pattern Filter
First we want to assess the value
of the MVI indicator as a pattern
filter. For this purpose we use an
objectively defined pattern and
see if we can pimp it with our
newly discovered oscillator. As
an objectively defined pattern we
chose a key reversal bar, defined
as a bar opening below the
previous bar’s close but closing
above the previous bar’s high.
In our back test we round up all
those key reversal days and have
a look at the average profit for
each day forward after such a
bar.
Figure 2 shows an average
profit of about three per cent, 30
days after a key reversal day was
taken. After 30 days the effect of
a key reversal day seems to wear
off. In the first five to seven days
the chart shows an average loss
never amounting to more than
one per cent.
When the key reversal days
get filtered with a Monest Value
indication on top, allowing only
those key reversal day entries
to be taken when they were
accompanied by an MVI less
than -4, the number of valid entry
signals drops by about 50 per
cent.
However, the average profit on
30 days almost doubles. This can
be seen on the chart in Figure 2.
Of course, adding an additional
criteria to the signals taken, can
never change anything about the
wearing off effect. After about
one month, the signal effect fades
away. What is more is that the
minor (average) adversity in the
beginning of a key reversal day
trade seems to be half in length
(only about three days instead of
up to seven).
System Filter
A key reversal day is just a
pattern. We see patterns
everywhere. We are evolutionary
designed to do this. In
evolutionary terms, it pays off
to assume a tiger where there
is not one (false positive). That
programming happened a really
long time ago, because even a
horse is scared of a garden hose
F1)
Different Entries vs. Undervalued MVI Entry
Buying while short-term undervalued (MVI <= 7) seems to pay off
against other entry strategies. All systems are compared to a random
entry system.
Source: www.chartmill.com
F2)
The Monest Value Indicator as a Pattern Add-on
Effect on the average return up to 50 days after a key reversal day when
only the key reversals are taken that are accompanied by an MVI < -8.
This demonstrates the Monest Value Indicator as a pattern catalyst.
Source: www.chartmill.com