Technical Analyst

nov/dec 2004
The publication for trading and investment professionals
www.technicalanalyst.co.uk
The January Barometer

An early predictor for stocks?
Outlook for CAD
Fine-tuning Fibonacci
Behavioural Finance
Long-term view
remains bearish
Identifying when
reversals are likely
Investor psychology
& the charts
London Stock Exchange Advertisement
WELCOME
In our last issue of the year, we look at the January Barometer as a means of predicting stock market performance for 2005. We also take a step into the world of
behavioural finance, a subject that has become an increasingly significant part of the
trading and investment vocabulary in recent years. We present an introduction to
commonly accepted terms and phrases used in behavioural finance and make some
attempt at linking investor psychology with TA. Back in the world of traditional technical analysis, Fibonacci is a subject familiar to most traders and is usually applied in
determining price retracement levels. This issue discusses how Fibonacci ratios can
be used along with conventional indicators to better identify market turning points and
enhance trading strategies. Finally, Bent crude continues to dominate market news so
a leading energy broker gives its view of where prices should head in 2005. We hope
you enjoy the magazine and find time to visit us at www.technicalanalyst.co.uk
Matthew Clements, Editor
CONTENTS 1 > FEATURES

Forecasting stock market direction for the
forthcoming year is the goal of every equity
analyst and trader. Does January's performance provide any guide to the year ahead?
Fibonacci revisited
Combining indicators and Fibonacci
retracements to identify market
turning points
Investor psychology
The key psychological biases that affect
investor behaviour and the charts.
2004 Clements Biss Economic Publications Limited. All rights reserved. Neither this publication nor any part
of it may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic,
mechanical, photocopying, recording or otherwise, without the prior permission of Clements Biss Economic
Publications Limited. While the publisher believes that all information contained in this publication was correct
at the time of going to press, they cannot accept liability for any errors or omissions that may appear or loss
suffered directly or indirectly by any reader as a result of any advertisement, editorial, photographs or other
material published in The Technical Analyst. No statement in this publication is to be considered as a
recommendation or solicitation to buy or sell securities or to provide investment, tax or legal advice. Readers
should be aware that this publication is not intended to replace the need to obtain professional advice in
relation to any topic discussed.
November/December 2004
NOV/DEC
>16
>28
>32
>>
THE TECHNICAL ANALYST
GET QUALIFIED IN TECHNICAL ANALYSIS

The Society of Technical Analysts (STA) represents and accredits
professional and private Technical Analysts operating in the UK
Originally established in the 1960s, the STA provides its members:
Education
Monthly lectures and regular teaching courses in technical analysis
Research
The STA Journal publishes research papers on TA techniques and approaches
Meetings
Provide members the opportunity to discuss technical approaches and markets
Representation The STA lobbies on behalf of analysts with data vendors, exchanges and regulators.
The STA represents the UK at the International Federation of Technical Analysts (IFTA)
Accreditation
The STA Diploma Exam is internationally recognised as a professional level qualification

in Technical Analysis
The next Diploma examination day is 22 April 2005 at the London School of Economics
For more information on how to join and what is involved in passing

the STA Diploma exam, visit our website at: www.sta-uk.org or call
To subscribe to The Technical Analyst magazine

post a copy of this form to:
The Technical Analyst
Unit 201
Panther House
38 Mount Pleasant
London WC1X 0AN
Or email subscriptions@technicalanalyst.co.uk,
providing the information requested in this form
SUBSCRIBE
us on +44 7000 710207
I/We would like to subscribe to

The Technical Analyst magazine for:
140 1 year subscription - UK
165 1 year subscription - Rest of World
Please tick as appropriate
Name:
Title:
Institution:
Address:
City/State:
Country:
Postcode:
Phone:
Email:
I attach a cheque payable to The Technical Analys
Please invoice me/my company
Alternatively, email subscriptions@technicalanalyst.co.uk,
providing the same information as requested above.
+44 (0)20 7833 1441
www.technicalanalyst.co.uk
CONTENTS 2 > REGULARS

Editor: Matthew Clements
Managing Editor: Jim Biss
Editorial Board:
Mikael Bask, Umea University, Sweden
Tai-Leung Terence Chong,
The Chinese University of Hong Kong
Wing-Keung Wong,
National University of Singapore
Marketing: Vanessa Green
Advertising Sales: Charles McCrostie
Design: Paul Simpson
The Technical Analyst is published by
Clements Biss Economic Publications Ltd,
Unit 201, Panther House,

38 Mount Pleasant, London WC1X 0AN
Tel: +44 (0)20 7833 1441
Web: www.technicalanalyst.co.uk
Email: editor@technicalanalyst.co.uk
INDUSTRY NEWS
Latest news, BIS survey and IFTA roundup
04
MARKET VIEWS
USD/CAD - the bears refuse to hibernate
Elliott Wave outlook for gold remains bullish
Outlook for USD/CHF
Brent crude at the crossroads
08
10
12
14
TECHNIQUES
Are exit strategies more important than entries?
Quantifying market deception with the Hikkake pattern
Fine-tuning Fibonacci
16
20
23
28
SUBSCRIPTIONS
Subscription rates (6 issues)
UK: 140 per annum
Rest of world: 165 per annum
For information, please contact:
subscriptions@technicalanalyst.co.uk
ADVERTISING
THE TECHNICAL ANALYST TALKS TO

Jorge Bolvar, chairman of the AEAT
SUBJECT MATTERS
Investor psychology - the thinking behind the charts
Recognizing patterns with fuzzy logic
For information, please contact:

advertising@technicalanalyst.co.uk
PRODUCTION
Art, design and typesetting by
all-Perception Ltd.
Printed by The Friary Press
ISSN(1742-8718)
31
32
36
SOFTWARE
CBOT Market Profile
40
BOOK REVIEW
New Market Mavericks by Geoff Cutmore
43
COMMITMENTS OF TRADERS REPORT

LONG-TERM TECHNICALS
TRAINING AND EVENTS DIARY
44
46
48
Industry News
eSIGNAL ADDS BRAZIL AND

MEXICO EXCHANGES
eSignal has announced the availability of eSignal 7.7, the latest version
of its 7 series software. This release
includes new features such as News
Manager and Symbol Lookup as well
as new data from certain Latin
American exchanges. "We are
expanding our marketplace to Latin
America with the addition of three
new exchanges. The News Manager
enhancements make it much easier
to find market-moving news and we

have added a function key feature
that offers one-click access to the
most-often-used commands," said
Chuck Thompson, President of
eSignal. The three new exchanges
offered through eSignal are the
Mexico Stock Exchange, the Bolsa
de Mercadorias & Futuros (Brazil)
and the Sao Paulo Stock Exchange.
OPEN INTEREST
RISES TO RECORD
LEVEL
Eurex has reported record

open interest on the
exchange for October. The
derivatives exchange
announced that total open
interest rose to 76.4 million
contracts, an increase of
22% on October 2003, the
highest level ever on the
exchange.
Trading highs have also
been recorded at the
Chicago Board of Trade
(CBOT). On Friday,
November 5th, 4,324,024
contracts were traded, the
second highest total volume for a single trading
sessions in the exchange's
history.
UPDATA RELEASES MARKETHAWK
Charting software provider, Updata,

has released MarketHawk which has
been developed to send market alerts
to mobile or handheld devices. As well
as providing handheld charting capabilities in conjunction with Updata
4
charts, MarketHawk can also send

SMS alerts notifying the user of specified market moves, stop-loss breaches
and keyword news stories.
MarketHawk is available in both
real-time or end-of-day versions. The
real-time version costs 57 per month
and is available for free trail at:
www.updata.co.uk.
HSBC LAUNCHES
CHARTING PACKAGE
ON FXALL
HSBC has launched a free online FX
charting package on the FXall website,
covering 15 major cross rates. The service at www.fxall.com is provided by UKbased charting provider, TraderMade.
The package allows users to chart moving averages, measure volatility, indicate
overbought and oversold activity, and
give warning signals of potential trend
reversals. HBSC is the first of FXall's 57
FX price-providing banks to offer technical analysis and charting.
Industry News
BIS 2004 REPORT:

SHARP INCREASE IN TURNOVER SINCE 2001
The Bank of International Settlements (BIS)

Triennial Survey of Foreign Exchange and Derivative
Market Activity has reported a dramatic upturn in
trading volumes from 2001 to 2004. Average daily
turnover in foreign exchange markets as reported by
52 central banks and monetary authorities around the
world rose to $1.9 trillion in April 2004, a 57%
increase over April 2001 at current exchange rates.
Total trading in FX and interest rate derivatives rose
to $1.22 trillion in swaps, options, FRAs and forwards. This figure is up from $0.575 trillion in April
of 2001.
A spokesman for the BIS told The Technical Analyst,
"Market commentary suggests both money managers
and hedge funds contributed to the increase in trading. This is in contrast to the 2001 report when
growth was reportedly driven mainly by pension
funds and insurance companies."
Trade in UK sterling has also increased relative to
other currencies with sterling/dollar rising from 11%
of total turnover in 2001 to 14% in 2004. According
to the BIS, "The presence of a clear trend for sterling
and favourable interest rate differentials appear to
have boosted trading in the currency, as well as that
in the Australian and New Zealand dollars and several emerging market currencies."
Trading in interest rate options for all currencies
rose especially sharply, up almost 6-fold in the three
years to 2004. The BIS says this was largely due to
interest rate differentials and uncertainty surrounding
future rate levels between 2001 and 2004.
Foreign exchange average daily turnover ($bn)

%
1992
1995
1998
2001
2004
Spot
394
494
(43%)
568
(40%)
387
(33%)
621
(35%)
Forwards
58
97
(9%)
128
(9%)
131
(11%)
208
(12%)
Swaps
324
546
(48%)
734
(51%)
656
(56%)
944
(53%)
Total
820
1,190
1,490
1,200
1,880
Turnover share by currency pair (%)

Currency pair
1995
1998
2001
2004
EUR/USD
30%
28%
USD/JPY
21%
18%
20%
17%
GBP/USD
7%
8%
11%
14%
USD/CHF
5%
5%
5%
4%
Industry News
IFTA CONFERENCE IN BRIEF
Marc Michiels with conference speakers
The annual IFTA conference held in Madrid in November gave market

practitioners an opportunity to listen to professional TAs from around
the world, including prominent personalities such as Murphy, Bollinger
and Pring. In this brief roundup, we present a few of the key findings
that testify to the ongoing development of technical analysis.
RSI TOPS POLL
MURPHYS OUTLOOK
Independent analyst Gerald

Butrimovitz presented a poll of
the top 20 favourite indicators.
The survey showed that 8 of the
top 10 are price based indicators
and only 2 are volume based.
The most popular reported indicator is the RSI. The poll included a total of 228 indicators, of
which 188 are price based and
40 volume based.
TA guru, John Murphy's talk

focused on the use of intermarket analysis, especially in
relation to the weakness in
the dollar continuing to support higher gold prices.
Murphy sees the Dow in a
bullish short-term correction
until year-end or April latest.
After that, its downhill for a

long-term bear market.
Murphy also advocated the
application of fundamental
news with technicals and the
use of contrary opinion in
making investment decisions.
As he put it, "When you
hear it on TV, it's too late".
Industry News
BLOOMBERG ON DEMARK
Trevor Neil, global head of technical analysis at Bloomberg, gave
a persuasive talk on the DeMark
Sequential Indicator. Bloomberg
is one of the principal providers
of DeMark and Neil provided
numerous market examples of
the Sequential Indicator in action.
Whilst DeMark is only around

50% reliable, Neil explained that
where it gives correct buy and sell
signals, these tend to precede
larger price moves. As such,
DeMark provides positive returns
when used for timing entry and
exit points.
BOLLINGERS BANDS
THE SENTIX
John Bollinger, not surprisingly, stuck to Bollinger

Bands and showed how they
could be applied to different
markets around the globe
with very minor modification. Broadly speaking
though, his research showed
that the settings on the
bands (2 standard deviations,
20-day MA) are still relevant
Manfred Huebner of Deka

Investment and Rolf Wetzer of
MEAG Asset Management presented the Sentix sentiment indicator. Introduced in 2001, the
German based survey of investor
opinion polls its 1500 members
every Friday regarding the short
and medium term outlook for 10
markets including stocks, bonds
and currencies. Sentix is available
via Bloomberg, Reuters and
Thomson.
today. Bollinger also touched

upon the idea of applying
the bands to the RSI oscillator. This technique, he said,
gives a more precise indication of when a market is
overbought or oversold
because it recognizes that
the OS and OB ranges shift
according to the condition
of the market.
NEAREST NEIGHBOURS
Jorge Bolvar, chairman of
the Spanish Association of
Technical Analysts, presented a paper called "Nearest
Neighbour Pattern
Recognition". NNPR aims to
find the closest price patterns from past data to the
most recent price action. If

the closest price patterns
generated similar price
action, then Bolvar says we
may be in front of a predictable situation. According
to Bolvar, the system, which
is rooted in chaos theory, has
achieved winning trades of

70-80% - in tests. This was
achieved without overfitting
or any assumptions about
which chart patterns are the
most effective.
Market Views
USD/CAD
THE BEARS REFUSE TO HIBERNATE
y early November, USD/CAD had depreciated by

approximately 24.5% since January 2002 when a secular high of 1.6194 was registered. In Figure 1, we
note the fact that USD/CAD had traced out a rising wedge
pattern and a potential 5-wave Elliott wave structure as
prices reached their peak (labelled 1-5). With bearish divergences also in place on the monthly momentum and slow
stochastic studies, the watershed moment for USD/CAD
took place in January 2003 when the pair registered a
monthly close below a 12-year support trendline at 1.5225.
This generated a bearish reversal of the long-term uptrend
that had been in effect since 1991 - resulting in one of the
largest one-year declines in USD/CAD on record from
1.5784 in January 2003 to 1.2683 in January 2004. After
forming a doji pattern in December 2003 as the momentum
Figure 1.
and slow stochastic studies reached 13-year oversold

extremes, a retracement phase commenced. Interestingly,
the correction halted right against the 38.2% Fibonacci
retracement of the January 2002-January 2004 move from
1.6194 to 1.2683 at 1.4024 (the high was 1.4001 in the
month of May).
by George Davis
Prices reaffirm long-term downtrend

The July monthly close below trendline support at 1.3396
reversed the medium-term corrective uptrend that had been
in effect since January - reinforcing the price failure against
the 38.2% Fibonacci retracement level. In addition, prices
have just registered a monthly close below a redrawn longterm support trendline at 1.2776 that is derived from the
cyclical low in 1991 and the reaction low in January of this
year. The September close below this level has reaffirmed
the long-term downtrend in USD/CAD that began in
January 2003 and indicates that the current impulsive 5th
wave of an Elliott wave structure has the scope to extend
lower (labelled A-E). Based on our bearish outlook, resistance at 1.2776, 1.3071 and 1.3446 is expected to cap rallies
and attract long-term selling pressure. Initial support is
located against the March 1993 reaction
low at 1.2402. A monthly close below
1.2402 would then target the descending
channel base at 1.1968, followed by a
double bottom from July-August 1992
at 1.1819.
From a Fibonacci standpoint, we note
that prices have already exceeded a
61.8% retracement of the 1991-2002
secular upmove from 1.1192 to 1.6194
at 1.3103. The 76.4% retracement level
at 1.2372 is now in view and a monthly
close below this level would then establish a long-term focus on the 1991 low
at 1.1192 based on the premise of a full
100% retracement. Although the
momentum study is displaying a bullish
monthly divergence, we point out that
such divergences can persist for multiple
monthly (even yearly) periods before
they are resolved - as was the case
between 1992-1994 and 1998-2002.
With the stochastics crossing bearishly
from neutral levels, additional downside
price momentum remains viable. Prices
will have to register a monthly close above the descending
channel top at 1.3446 in order to reverse the current longterm downtrend and eliminate our bearish outlook - a
development that would target 1.4001 and 1.4188 respectively (the high from June 2004 and the double top from
July-August 2003).
Market Views
WHILE THE LONG-TERM OUTLOOK IS

CLEARLY BEARISH, THE MEDIUM-TERM
PERSPECTIVE PRESENTS A MORE
TEMPERED BEARISH VIEW.
Figure 2.
Retracements to provide selling opportunity

While the long-term outlook is clearly bearish, the mediumterm perspective presents a more tempered bearish view.
More specifically, the daily studies are lingering in oversold
territory as a series of bullish USD divergences remain in
place (see Figure 2). Recent IMM data also indicates that
net long CAD positions are at historical highs (a record
49,386 contracts were registered on October 12), raising the
increased possibility of a price correction from a positional
standpoint. With prices piercing below the base of both the
major and minor descending channel patterns, initial support is located at 1.2175, 1.2115 and 1.2007. These levels
are likely to attract some buying pressure as the positioning
of the studies indicates that the current impulsive thrust
lower may not be sustainable.
The minor descending channel base at 1.2334 serves as a
key pivot point with regard to a potential price correction.
Specifically, a daily close above 1.2334 would cause prices to
re-enter the channel and confirm the

posture of our studies. This offers scope
for a more sustained medium-term price
correction that would target resistance at
1.2420, the minor descending channel
top at 1.2535 and the January low at
1.2684. We expect such corrections to
attract medium-term selling pressure
between 1.2420/1.2684 for a sustained
return below 1.2334. Only a daily close
above the major descending channel top
at 1.2941 would abort our bearish
stance, thereby reversing the mediumterm downtrend and projecting a greater
retracement to the July high at 1.3383.
Both the medium and long-term
charts indicate that USD/CAD remains
within an established downtrend.
Retracements that arise due to overextended medium-term valuation metrics
are expected to present a selling opportunity between 1.2420/1.2684 for a
more sustainable return move below
1.2334. Long-term price targets are located at 1.1968/1.1819. We expect
USD/CAD to end the year near 1.2300 as price retracements are met with selling pressure, reflecting our overall
bearish outlook.
George Davis is chief technical analyst with Royal

Bank of Canada Capital Markets in Toronto.
Market Views
ELLIOTT WAVE OUTLOOK FOR

GOLD REMAINS BULLISH
by Craig Ferguson
he long-term outlook for gold continues to suggest

significant further upside for prices. While Figure 1
only looks at prices for the last 7 years, a look at a
chart going back to the peak in prices at $850 illustrates that
the entire ABC correction from that high completed itself
at the August 1999 low of $252. This suggests that the
decade- long outlook is for a return to $850 or higher.
Over the medium term, the rally that started at $252 in
August 1999, completing in October at $340 constitutes
wave 1 in the expected decade-long bull market. The
retracement back to $254 in February 2001 was wave 2 and
now prices are in the middle stages of wave 3. Ultimately a
5 wave advance will unfold taking prices towards the
December 1987 peak at $503, part of the eventual decade
long move back above $850. The alternate view is that the
$850-$252 decline, rather than being the end of the correction, is actually the first (A) wave of the correction. Even
so, prices would still rise to $503-$850 over the multiple
year timeframe before moving back to $350-$250. Either
way, the outlook for the medium-term remains bullish.
Over the short-term, the issue is whether prices will continue their current wave 3 advance through $430 towards
$450-$475 or slump back into the years range between
$430-$370. The Dow outlook suggests that stocks and the
USD could still rally for another few months before slumping in 2005. In this case it's possible that gold would remain
range bound for another 2-3 months, even retesting its
downside extreme at $370 before heading higher. This view
is conditional not just on US stock market gains, but also
on the USD remaining range bound. Gold prices now
Figure 1.
10
Market Views
ULTIMATELY A 5 WAVE ADVANCE WILL

UNFOLD TAKING PRICES TOWARDS THE
DECEMBER 1987 PEAK AT $503, PART OF
THE EVENTUAL DECADE LONG MOVE
BACK ABOVE $850.
Figure 2.
appear to be consolidating their breakout above $430, and

while they remain above $420-415, could easily extend their
recent gains.
As always when we look at gold prices, we seek supporting evidence from gold stocks, and as such we look at the
Amex Gold Bugs Index, or HUI (Figure 2). The HUI is
very interesting at this juncture. The long-term wave pattern
shows a clear low in November 2000 around the 35.30 level
and a substantial rally to 258.60 on December 7 2003. Since
then the index has been in a substantial range, which also
made a low at 163.80 in May of this year. The long-term
outlook calls for the index to move towards at least 350400, in sympathy with gold prices. Along those lines, it is
possible while the HUI remains below 258, that a 5 wave
advance from 35.30 is now complete and that the index is

within a protracted 1-2 year range. Under this view, the
HUI will remain within its 258-163 range for another few
months and could decline to retest the 163 level before
heading to new highs above 260. However, it is also possible that rather than 5 waves up being complete, the HUI
would still be within wave 3 heading to 300-350. Even
under this exceedingly bullish scenario, it remains possible
for the HUI to remain within the 258-163 range for another
few months before moving topside again.
Craig Ferguson is technical analyst with ANZ
Investment Bank.
11
Market Views
OUTLOOK FOR USD/CHF

by Patrick Dyess
ith bearish sentiment surrounding the dollar and

heightened geopolitical factors supplying an
increased demand for the safe-haven currency,
the trend for USD/CHF appears quite clear. Yet early this
year the dollar gained some traction, taking a brief pause
from the trend. The move brought USD/CHF to the upper
reaches of the channel dragging the 26-week SMA to a near
crossover of the 52-week in the process. The reversal was
short-lived, however, and the pair recently headed back
down dramatically, breaking a 6-month old support line. For
FX observers around the world the event was rather
momentous.
For a clearer idea of where USD/CHF may go from here,
it's worthwhile looking at support and resistance levels. A
common technique for predicting key support and resistance levels applies the Golden Ratio (phi) to a historically
significant period of price action, for example the high and

low of price during a significant market event. In the case
of USD/CHF, the reversal of the previous trend came
shortly after the burst of the technology bubble and the
beginning of the Fed's easing cycle in September 2000. It
concluded once US monetary policy reached its ultra
accommodative period in May 2001, creating a double top
(see Figure 1).
By measuring the gap of the formation we can project
future support and resistance at the reciprocal of phi
(0.618), the 1:1 and of course the Golden Ratio (1.618). In
this instance, the significance of these levels is marked by
the stall at 1.4400/50 (reciprocal of phi) and the serpentinelike price action around the 1:1 projection at 1.3500/50 in
March of 2003. Later still, the Golden Ratio provided the
base of the summer's range near 1.2050/2100, providing
Figure 1.
12
Market Views
...THE GOLDEN RATIO PROVIDED

THE BASE OF THE SUMMER'S
RANGE NEAR 1.2050/.2100.
Figure 2.
confirmation of this measure.

As USD/CHF approaches any given level, one can look at
a shorter-term chart to look for confirmation of a reversal,
such as that shown in Figure 2. Studies such as the stochastic oscillator, RSI and MACD also provide important signals
of a shift in consensus. Using this method we can see that a
move higher from present levels will likely be resisted at the
former support line near 1.2000/2050 and above that where
the channel upper band meets with the 100/200-day SMA
(1.2475/2595).
A move lower on the other hand brings an alternate scenario. The channel centerline and first Fibonacci level intersect early 2005, creating a wedge pattern. If the price compresses into the apex of this wedge we could see another
explosive move much like the one that broke down the Q1Q2 2004 range. This is just one of many potential scenarios. Moves below 1.1485 could likely bring the 1.1135 level
into the picture.
We can also use phi to estimate when such pivotal
moments can be expected, in much the same way as we can
project important price levels. Using the same starting
point, this places the reciprocal on 19th November and the
1:1 in the first week of December. The golden ratio can be
expected at January 5th 2005. Interestingly, this is also when
the wedge pattern comes to an apex.
Patrick Dyess is technical strategist at RefcoNews.
13
Market Views
BRENT CRUDE AT THE CROSSROADS

by Michael Davies
rent crude has rocketed to all time highs at over $50

a barrel recently, $11 above the previous high set in
October 1990 when Iraq invaded Kuwait. A small
proportion of market analysts looking at the fundamentals
believe supply is so tight now that prices could go as high
as a $100 in the years ahead. A larger proportion believe
these recent highs have prompted concern that oil may be
dragging back the world economy, leading to a fall in
demand for oil and thus lower Brent prices. A look at the
Brent crude daily chart (Figure 1) shows that the market has
broken through the short-term uptrend and is currently
held above support at $40. However, it appears that
November is a crossroads with a case for a move in either
direction that could see Brent in completely different price

ranges three months from now.
The pattern of the short-term uptrend in the Brent crude
daily chart shows how similar wave A is to wave B. The
market rose by $11.84 between 30th June and 20th August
2004 (Wave A) and by $11.95 between 16th September and
27th October (Wave B). If the uptrend continued, the
beginning of February 2005 would have seen crude almost
$12.00 higher, near $57.50 a barrel. There would be
grounds for this move if President Bush continues his policy of filling emergency oil reserves and China's demand
holds up. However, there are signs that the market may
have reached its peak.
Figure 1.
14
Market Views
There are two possible signals that may confirm the market top has been reached both of which can be seen on the
daily chart in Figure 1. Firstly, wave B shows a potential
double top with a difference of less than 1% between
peaks. However, the low height of the peaks and the small
distance between them indicate perhaps only a weak reversal signal. The second signal on the daily chart comes from
the RSI which indicates divergence from the recent highs.
In this situation it appears the bulls are losing their grip on
the market, prices are rising only as a result of inertia, and
the bears are ready to take control again. The divergence
appears to be a class B bearish divergence, illustrated by
prices making a double top with an oscillator tracing a

lower second top. There may be a correction higher in the
short-term but in the long-term (Figure 2) the market looks
set to head lower returning to its long-term trendline at
around the $40.00 to $35.00 area over the next 3 months.
What is certain is that $51.95 is a key resistance level preventing the continuation of the short-term uptrend to
$57.50 a barrel.
Michael Davies is an analyst at commodities and
financial futures broker, Sucden.
Figure 2.
15
Techniques
DOES JANUARY SET THE
he January barometer is a market indicator that says if the

market rises in January, then it
will also rise during the year as a whole.
Similarly, if it falls in January, then the
year will also end down. Therefore
January can signal or confirm whether
to stay in or out of the market for the
rest of the year.
Table 1 lists the percentage change
for the Dow in each January and for the
year as a whole from 1960 to 2004. The
change for months that ended up are in
white and those that ended down are in
red. In the 44 years since 1960 (2004
not included) the barometer worked on
36 occasions, an impressive success rate
of 82%. Based on this simple analysis,
the barometer does appear to have
16
strong predictive power in gauging the

market direction of the year ahead.
A closer look at the data reveals that
the barometer has a slightly higher success rate for up years than for down
years. Since 1960, 30 years have been
bull markets of which 26 (87%) were
predicted by the barometer. Whereas of
the 14 bear market years, only 71% or
10 were anticipated. This is a common
result for US indices perhaps because
down years are often caused by unforeseen events such as wars, shock election
results and recessions while up years
are often the result of a continuation of
a bullish long-term trend.
Based on this simple analysis, the
barometer does appear to have strong
predictive power in gauging the market
direction of the year ahead.

Lame ducks
According to Jeff Hirsch of the Stock
Traders Almanac, the basis for
January's predictive capacity comes
from the fact that so many important
events occur in the month. The reasons
behind this, he says, are largely historical and can be explained to some extent
by the passage of the 20th "Lame
Duck" Amendment to the US
Constitution in 1934. Prior to this,
newly elected senators and representatives did not take office until December
of the following year, except when new
Presidents
were
inaugurated.
Consequently, defeated congressmen
stayed in Congress for all of the follow-
Techniques
THE JANUARY BAROMETER
TONE FOR THE YEAR AHEAD?

Changes in investor sentiment and the
release of certain fundamental news in
January could mean that January's stock
market performance is an important
leading indicator for the rest of the year.
ing session and, as such, were known as

"lame ducks." Since 1934, Congress
convenes in the first week of January
and includes those members newly
elected the previous November.
Inauguration day was also moved from
March 4 to January 20. As a result, several key political events have shifted
into January that impact on the US
economy and stock markets. The most
notable is the president's State of the
Union address that presents the annual
budget and sets forth economic goals
and priorities. Hirsch adds that other
factors also contribute to the effect,
namely the large amount of cash flowing into the market in January as
investors make fresh decisions about
the year ahead.
Hirsch says that the years that saw the

barometer fail can be attributed to market distortion caused by heightened
uncertainty, usually the result of US
involvement in military conflict. A look
at the events of these years appears to
bear this out. For example, the barometer performed poorly during the late
1960s (Vietnam) and after 2001
(September 11th and Iraq). However,
the barometer has attracted criticism
for the inclusion of January itself in the
performance of the year as a whole.
Surely, if January experiences a large
move in either direction then this will
contribute to the year's overall move
thereby greatly diminishing the barometer's true predictive power? To overcome this, the performance of the
Dow over just the next 11 months is

compared. For this, the barometer does
work less well, predicting the direction
of the Dow on only 31 out of 44 years,
or 70% of the time.
Applying a 1% threshold
In some years, such as 1998, the move
in January is so small as to be inconsequential. One way of eliminating these
years is to apply a 1% threshold. As
such, any January move of less than
1%, either up or down, is ignored. This
would effectively omit the years where
the barometer has only a 33% success
rate. In total, six years fall below the
threshold level of 1% between 1960
and 2004. Once the threshold is
applied, the success rate of the
17
Techniques
% change in the Dow

Year
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
Table 1.
18
January
change
-8.4
5.2
-4.2
4.8
2.9
3.3
1.6
8.1
-5.5
0.2
-7.0
3.6
1.4
-2.1
0.6
14.3
14.6
-5.0
-7.3
4.2
4.4
-1.8
11.0
2.8
-3.0
6.2
1.6
13.8
1.0
8.0
-5.9
3.9
1.7
0.3
6.0
0.3
5.4
5.7
0.0
1.9
-4.8
0.9
-1.0
-3.6
0.33
Year
change
11-month
change
-9.3
18.7
-10.8
17.0
14.6
10.9
-18.9
15.1
4.3
-15.3
4.9
6.1
14.6
-16.6
-27.6
38.3
18.0
-17.3
-3.1
4.2
14.9
-9.2
19.7
20.3
-3.7
27.6
22.6
2.3
11.9
26.9
-4.3
20.3
4.2
13.7
2.1
33.5
26.0
22.6
16.1
25.2
-6.2
-7.1
-16.8
25.3
?
-1.0
12.8
-6.9
11.7
11.3
7.3
-20.1
6.5
10.4
-15.4
12.8
2.4
13.1
-14.8
-28.0
21.0
3.0
-13.0
4.5
0
10.0
-7.6
7.8
17.1
-0.7
20.2
20.7
-10.1
10.8
17.5
1.7
15.8
2.4
13.4
-3.6
33.1
19.5
16.7
16.1
22.8
-1.4
-7.9
-15.9
29.8
?
Techniques
THE BASIS FOR

JANUARY'S PREDICTIVE
CAPACITY COMES FROM
THE FACT THAT SO MANY
IMPORTANT EVENTS
OCCUR IN THE MONTH.
- JEFF HIRSCH
barometer rises to 34 years out of 38,
or 89%. For the next 11 months'
change the barometer works on 29
years out of 44 when the threshold is
applied.
Is January unique?
In a recent paper, Lawrence Brown of
Georgia State University argues that
much of the research into the barometer fails to show that January is any better as a predictor than other months.
Consequently, he measured the relative
performance of January against the
other calendar months in predicting the
performance of stocks over the following 11 months. He looked at all New
York Stock Exchange listed stocks
from 1941 to 2003 and his results show
that January does indeed have the
greatest predictive power for the next
11 months. However, he also found
that stock market investments made on
the basis of the barometer at the end of
January don't necessarily provide the
best returns over the year when compared with other months. Mark W.
Riepe, an analyst with Charles Schwab
attaches less importance to January specific events in explaining the barometer,
"I think what's happening is that it isn't
January per se, but rather that if there
are favourable conditions for stocks in
a given month, then those conditions
have tended to persist."
What about 2004? The Dow needs to
close above 10,453 to work this year. At
the time of writing the barometer looks
set to notch up another success.
The January Effect

The well known January Effect (not to be confused
with the January Barometer) refers to the tendency of
small capitalisation stocks to rally during January after
a sell-off in December. There appears to be no consensus of opinion as to the cause of the effect
although many ideas have been put forward. The most
frequently cited include:
Tax loss selling - This is the most commonly excepted explanation. Year-end tax loss selling in December
causes stocks to bounce back in January as individuals sell their small cap stocks at year-end for tax purposes.
January optimism - Positive sentiment that the market will have a good year prompts heightened buying.
Earnings - Research conducted by the University of
Southern California has concluded that corporate
earnings data released in January plays a significant
part in explaining the effect. They observe that most of
the gains occur in the first five trading days of January
but that late January can also see above average
stock returns. This may be explained by the tendency
of many corporate earnings announcements in
January to be positive. Firms are more willing to report
good results as soon as possible and because of the
large number of firms that have their year-end in
December, and so report annual results in January.
Year-end bonuses and dividends - Used by individuals to invest in the stock market after end-of-year festivities.
Is the January Effect still valid? Mark W. Riepe of
Charles Schwab says that increased awareness of the
effect appears to have diminished its validity over the
past four or five years for all but the smallest companies. Moreover, any attempt to exploit whatever effect
remains is hampered by transaction costs.
19
Techniques
ARE EXITS STRATEGIES MORE

IMPORTANT THAN ENTRIES? by Simon Brown
Despite unanimous agreement that sound exit strategies are
essential components of trading systems, traders regularly
debate whether disciplined exit strategies alone can produce
consistent profitability within a rule-based trading framework. Exit
strategies can have a great impact on both profitability and risk.
A well-positioned exit leaves room for error on a trade entry,
allowing a position to capture the maximum profit while avoiding getting stopped out too soon. But can exit strategies possibly
be as important, or more important, than entry strategies?
Establishing a test plan

Can exit criteria alone form the basis of
a successful trading system regardless
of the trade-entry technique? To investigate this theory, research is carried out
on an existing system designed for trading stocks from the S&P 100 index
(OEX). To conduct a meaningful test
of the exit-strategy, it is important to
establish a benchmark system for comparison.
Benchmark system
To keep things simple, it is important to
start with a trading system containing a
limited number of variable parameters.
The logic of the benchmark system
takes advantage of the large trading
ranges found in individual stock prices,
and the observation that stock prices
often bounce back from sharply over-
Testing approach
1.
2.
3.
4.
5.
Take a profitable system with an

existing position exit strategy
Collect the test results that can
be used as benchmark data
Remove the system's trade-entry
rule and replace it with a
random-entry technique
Test this system and compare
the results to the benchmark
system data
Finally, conduct additional runs
with the random entry technique
to tests the robustness of
the conclusion
Figure 1.
20
sold conditions.
In Figure 1, the horizontal lines
drawn above and below the market
highs and lows underscore the size of
the range established by Oracle
(ORCL) in recent months. They also
pinpoint areas at which the market had
become oversold or overbought, which
subsequently resulted in a rapid price
Techniques
reversal. Accurately identifying these

overbought and oversold areas in
advance would give a system a great
edge.
The following examples are based on
a reversal system that buys stocks at
oversold levels. These levels are defined
as a price move that is a certain number
of standard deviations from a longerterm moving average - 2.5 standard
deviations from the 90-day simple
moving average.
In a normal distribution, prices would
distribute about the mean price (as
shown in Figure 2), with 68% being
within one standard deviation of the
mean and 95.5% being within two standard deviations. If prices move to an
extreme - say, beyond two standard
deviations of the mean - it suggests the
market is overstretched and has the
potential to correct or reverse.
Plotting these standard deviation lines
above and below a moving average of
price on a chart results in a picture of
the moving bell curve as shown in
Figure 3. As prices reach the standard
deviation lines they often reverse quite
powerfully.
The rules for the benchmark
(long-only) system are:
1.
2.
3.
Entry: Buy if the price reaches

2.5 standard deviations below
the mean.
Exit: Sell half the position if
price goes up by 10% from the
entry price and sell the remainder of the position if the price
goes up by 20% from the entry
price.
Stop-loss: Exit the entire position if the stock price drops
10% below the entry price.
Testing the benchmark system

The system was tested on Boise
Cascade Co. (BCC) because the stock's
net performance over the test period
was negative (see Figure 4.). For the
purposes of this research we do not
want to use data with an inherent
Standard Deviations
68.3%
95.5%
99.7%
Figure 2.
upside bias because that could artificially favor the random-entry approach.
Table 1 shows the results of this system
from 1995 to the present, allocating
$5,000 per trade.
Overall, the system's statistics are
favorable. There are 60% winning
trades and the profit factor is greater
than 2. This is a good result and is at
the level we would require for a tradable system. One important point to
bear in mind is that the system, which
only trades from the long side, was able
to accomplish this over a period when
the stock price moved lower.
Replacing entry criteria with
random entry
Now the real analysis can begin: replacing the existing entry rule of the bench-
mark system with random entry signals.

This makes it possible to compare
results and establish whether the existing exit strategy by itself can maintain
profitability.
A random number generator was
used to create random trade entries. To
validate the comparison, random signals were generated at the same daily
frequency as the benchmark system of
about 10 trades over the test period. We
repeatedly tested the system using the
random entry and averaged the results
to get the data displayed in Table 1.
The random-entry system was also
profitable and had a higher percentage
of winning trades than the benchmark,
although total profit was reduced. The
profit factor of four on the random
entry version is considerably better
than the original version of the system.
Although this is only one test, these
results are worthy of further research.
By repeating the test many times over
for other stocks and using a random
number generator for entry signals, a
larger and more significant data set can
be collected and analyzed in greater
depth.
It should be noted that the exit criteria of the benchmark system tests were
fixed at 10% stop loss and profit
targets of 10% and 20%. These
Figure 3.
21
Techniques
Long Trades System
Long Trades Random
$28,012.29
10
$2,801.23
7.11%
6
$48,958.90
$8,159 .82
18.88%
4
$20,946.62
$5,236.65
10.55%
2.34
$16,178.43
8
$2,022.30
8.10%
6
$21,556.83
$3,592.80
14.40%
2
- $5,378.40
- $2,689.20
-10.80%
4.01
Net Profit
All Trades
Avg Profit/Loss
Avg Profit/Loss %
Winning Trades
Gross Profit
Avg Profit
Avg Profit %
Losing Trades
Gross Loss
Avg Loss
Avg Loss %
Profit Factor
Buy & Hold

- $6,876.54
1
- $6,876.54
-13.84%
0
$0.00
$0.00
0.0%
1
- $6,876.54
- $6,876.54
-13.84%
N/A
Table 1.
parameter levels were established during the development of the benchmark

system itself and were the result of
choosing the median value from the
range of profitable values for each
parameter.
Surviving randomness
The analysis demonstrates that position
management through profit targets and

money management stops can generate
a consistently profitable system even
when entry criteria are random. So can
disciplined exit strategies produce consistent profitability on their own? The
answer is yes and by focusing more on
exit strategies within a rule-based trading system, we can attain a positive
impact on both profitability and trade

risk.
Simon Brown is technical analyst at

Quantigma. www.quantigma.net
PROFIT TARGETS AND

MONEY MANAGEMENT
STOPS CAN GENERATE
A CONSISTENTLY
PROFITABLE SYSTEM
EVEN WHEN ENTRY
CRITERIA ARE
RANDOM.
Figure 4.
22
Techniques
QUANTIFYING MARKET DECEPTION

WITH THE HIKKAKE PATTERN
by Daniel L. Chesler
A savvy old timer once described Wall Street as "the only place
where they put prices up when they are having a sale." Indeed,
deception is a feature of most competitive fields including politics,
biological systems, financial markets and sports. Take for instance
the dummy move commonly used in football. The idea is to move
in one direction, thus unbalancing your opponent, before moving
away quickly in a different direction. The hikkake pattern is the market's version of the dummy move. Hikkake is a Japanese verb
meaning to trap or to ensnare. In western terminology, the proper
name for this pattern would be inside day false breakout.
Background
The pattern concept is that of a brief
pause in market action as defined by a
decrement in volatility, followed by a
false directional move. Volatility is
measured simply as the (hourly, daily, or
weekly, etc.) high to low range.
Academics might say the hikkake plays
on the short-term mean reversion
properties of markets, by identifying
situations where prices have been
stretched past their short-term equilibrium value. Momentum is generated in
the early stages of the reversal path,
causing prices to return to and eventually through their prior equilibrium
level. Technicians on the other hand,
should recognize the hikkake pattern as
a quantified compressed-time version
of the traditional "shake-out" pattern
(see Figure 1).
The rationale for why the hikkake
concept should have any effectiveness
may be due to the behavioral tendencies of small traders. Research by
Goetzmann and Massa suggest that

individuals (i.e., the public) are predisposed to "chasing" performance on a
short-term basis, and are equally predisposed to reversing their decisions when
the market moves against them. In
other words, individuals tend to be
short-term performance trend followers. In the case of the hikkake pattern,
perhaps it is the initial breakout that
attracts smaller participants into the
market. Another source of buying and

selling comes from technicians and system traders. Buying and selling around
predefined narrow ranges is an
extremely popular concept that goes
back at least to the early 1900s (see
Figure 2). In fact, strategies based on
this idea gained in popularity following
articles and books written by fund
manager Toby Crabel in the late 1980s.
However, if prices fail to gener-
R al l y To
New H i ghs
False Break
LONG BASE WITH SHAKE-OUT MOVE

Figure 1. This graphic, from William Jiller's classic 1962 book on charting, illustrates the traditional style "shake-out" pattern, which has also been described by authors such as
Schabacker, Wyckoff, and Edward & Magee. Hikkake patterns differ from this traditional
"false move" schematic primarily in respect to time.
23
Techniques
Figure 2. This graphic illustrates a technique for trading based on the narrow-range phenomenon (volatility expansion-contraction). From A. W. Wetsel's (Wetsel Market Bureau, Inc.) A
Course in Trading, 1933, with permission from Donald Mack.
ate the expected trending behavior following a breakout, a potential pool of

"trapped" new commitments is created.
Increased buying or selling pressure
from the unwinding of these losing
positions is one possible reason why
prices continue moving in the opposite
direction of a failed breakout. Another
possible source of fuel propelling the
market after a false breakout is the
opening of new positions in the direction of the true trend. For example,
many popular breakout strategies advocate placing stop and reverse orders
after an entry.
Description
The basic hikkake pattern is formed by
the combination of two price bars (two
hourly bars, two daily bars, two weekly
bars, etc.). The first bar must be an
"inside" bar, which is defined as any bar
that is completely encapsulated by the
previous bar's range. The second bar in
the pattern must have both a higher
high and a higher low than the previous
(inside) bar for a bearish hikkake set up,
or a lower low and a lower high than
the previous (inside) bar for a bullish
hikkake set up. Note that the hikkake
24
pattern ignores the positions of the

open and close, known as the "real
body" in candlestick terminology. This
is not atypical; other traditional candlestick patterns such as tweezers, hanging-man lines and hammers also ignore
the open to close relationship.
Once a hikkake pattern has formed,
the pattern is confirmed after prices
return up through the high of the
inside bar (for a bullish set up) or down
through the low of the inside bar (for a
bearish set up). Confirmation also
serves as the entry trigger when using
the pattern for trading purposes.
Normally I look for confirmation to
occur within three bars following the
initial, two-bar hikkake pattern. Upon
entering a position, risk is defined by
using the highest high (for shorts) or
lowest low (for longs) within the pattern as a stop out point. This is not necessarily the low (or high) of the false
move following the inside bar, since the
lowest low (or highest high) can also
occur anytime within the three bar period allotted prior to entry. The test
results presented in this article are
based strictly on the foregoing entry
and stop placement rules.
Examples
Unlike most price patterns that fall into
either "continuation" or "reversal" categories, the standard hikkake pattern
plays both roles equally well depending
on where it occurs within a trend. In
order to help the reader understand
how one pattern can play two very different roles, bullish chart examples will
be presented. All charts are from the
current year. Bearish examples will not
be presented due to space, but are mirror opposites of the bullish examples.
In April, June bond futures were in a
powerful downtrend. The market was
making new multi-month lows, longterm moving averages were pointing
down, and directional movement studies were showing that the strength of
the downtrend was actually increasing
(these indicators are not displayed for
sake of clarity). On April 26 an inside
day formed, giving us the first half of a
potential hikkake pattern (leftmost
point 1 in Figure 3). The next day
(point 2), a bar with both a higher high
and a higher low formed, completing
the
bearish
hikkake
setup.
Confirmation came quickly, on April
28, as the market traded below the low
of the inside day. A second bearish
hikkake pattern formed between April
30 and May 3, and was confirmed when
the market fell below the low of the
inside bar on May 4. Both of these
examples were found in the context of
an existing downtrend and demonstrate
how a bearish hikkake pattern functions in a continuation role.
Next we will look at an example of a
bearish hikkake functioning as a reversal pattern. The EUR/USD currency
pair peaked in early 2004. After correcting down into May, the EUR/USD
Techniques
began a gradual, low-momentum

recovery. On July 15 an inside bar
formed (point 1, Figure 4). The next
day, July 16, the market formed a higher high and a higher low (point 2), fulfilling the requirements for a bearish
hikkake. Confirmation came within two
bars, on July 20, as the market fell
below the low of the inside day.
Testing method and results
Tests were not aimed at determining
whether the hikkake has any value as a
stand alone system, but rather to find
out how often confirmed hikkake patterns represented valid false breakouts.
The ensuing moves were counted and
categorized by their magnitude. This
information should tell us whether the
hikkake pattern adds value as an analytical tool and can be used to guide our
entry and exit points.
For each valid hikkake pattern that is
confirmed within three bars following
pattern formation, a trade is entered
either long or short. The amount of
capital risked on each trade is defined
as the difference between the highest
high (or lowest low) of the confirmed
pattern and the entry price (either the
low or the high of the inside bar).
The test uses a static protective stop
that remains in its original location and
does not trail up or down as the trade
matures. The test then counted how
many times the ensuing move equaled:
Figure 3.
1.
2.
3.
4.
Figure 4.
Less than 100% of initial risk

before being stopped out.
100% to 200% of initial risk
200% to 300% of initial risk
Greater than 300% of initial risk
25
Techniques
RATIO
LONG & SHORT COMBINED STATISTICS
200+300
PERIOD
MARKET
STOPPED
100%-200%
200%-300%
>300%
WIN/LOSS
VS 100
01/02/90 - 08/15/04
DJI
64
21
14
42
1.20
2.72
01/02/80 - 08/15/04
GOLD
131
52
32
88
1.31
2.33
07/01/94 - 08/15/04
RUSSELL 2K
22
15
26
2.18
2.35
01/01/90 - 08/15/04
BONDS
60
26
48
1.35
2.17
01/05/98 - 08/15/04
USDJPY
22
10
18
1.64
3.70
01/05/98 - 08/15/04
EURUSD
28
12
19
1.25
2.02
04/30/90 - 08/15/04
NAT GAS
37
27
10
38
2.03
1.85
08/14/84 - 08/15/04
CRUDE
68
32
18
45
1.40
2.01
01/02/90 - 08/15/04
GE
44
17
23
1.23
2.25
01/02/90 - 08/15/04
MSFT
41
13
13
32
1.41
3.57
Table 1.
This information should give us some

idea as to the average magnitude of
moves following a confirmed pattern
and gives us some basis for setting
profit targets and for moving up stops
on open positions.
Markets tested include the Dow Jones
Industrial Index, Russell 2000 Index,
CBOT Bonds, COMEX Gold,
NYMEX Crude and Natural Gas,
USD/JPY and EUR/USD currency
pairs, and individual equities General
Electric and Microsoft. Test results do
Figure 5.
26
not include trading costs or slippage.

TradeStation daily data was used for all
tests. Continuous contracts were used
in the case of futures markets, which
may or may not accurately reflect the
results of testing on individual futures
contract data.
The results of the test show that over
fairly lengthy periods, ranging from
eight years in the currency pairs to
twenty-four years in gold, the hikkake
pattern produced an average win loss
ratio (number of winning trades versus
stopped out trades) of slightly better
than 1:1.
Table 1 shows the combined results
of all trades long and short. However,
we note from data compiled during
testing, that winning versus losing
trades for some markets faired much
better than the average combined performance, such as long natural gas
(3.2:1), short USD/JPY (3:1) and short
Russell 2000 (2.6:1). Less impressive
markets included short Dow Jones
Industrials (0.9:1), short bonds (1:1),
short EUR/USD (0.9:1) and long
USD/JPY (1:1). (A full breakdown of
individual long and short results is

available from the author).
Research ideas
Due to its simplicity, the hikkake pattern lends itself well to adaptations.
One version, which I have found effective at identifying trend reversals, takes
the standard pattern and applies the
following set of requirements to the
bar immediately preceding the "inside"
bar:
1.
The bar must close at the top of
its range (for top reversals) or
the low of its range (for bottom
reversals).
2.
The bar's range must be less
than the range of the previous
bar.
Note that since this version functions
primarily as a trend reversal pattern, it
occurs less frequently in the data than
the standard hikkake pattern.
The hikkake pattern is not limited to
traditional bar or candle charts. Point
and figure and tick-based charts, which
use a variable time axis, also exhibit
Techniques
UNLIKE MOST PRICE PATTERNS

THAT FALL INTO EITHER CONTINUATION OR REVERSAL
CATEGORIES, THE STANDARD
HIKKAKE PATTERN PLAYS BOTH
ROLES EQUALLY WELL .
hikkake patterns. Figure 5 displays an

example of both a bearish and a bullish
hikkake reversal pattern in the
EUR/USD, from May and September
respectively.
Final thoughts
I have a prejudice that says most technical indicators and methods are simply
different ways of reflecting the same,
or very similar, information. The common threads that bind indicators, patterns and methods are therefore more
interesting to me than are their apparent differences. In my own work, I have
always attempted to build models that
isolate these common elements.
My first attempt at this resulted in a
less subjective way of identifying classical chart patterns by distilling patterns
into separate volatility and cyclic components (see references). By compressing the traditional, detail-rich view of
classical chart patterns down to just
two main components, I discovered a
way to increase the method's robustness and improve its ability to generalize for unseen cases.
What patterns and concepts were dis-
tilled in order to arrive at the hikkake

pattern? Does the hikkake subsume the
minutia of other approaches? Earlier in
this article it was shown how the
hikkake concept overlaps with traditional shakeout patterns. Elliotticians
should find much in common between
hikkake reversal patterns and fifth-wave
terminations. Lastly, oscillator enthusiasts will find that momentum divergences on a lower time frame chart will
often correspond with hikkake reversals on a higher time frame.
In sum, chart patterns put traders in
a position to capture outsized moves,
while giving a defined structure for setting risk parameters and exit targets.
Patterns help us develop a case for
either a bearish or bullish outlook in
conjunction with other inputs. Chart
patterns add context that statistics do
not capture. The hikkake pattern serves
all of these purposes.
Daniel L. Chesler, CMT, CTA, is a
charter member of the American
Association
of
Professional
Technical Analysts (AAPTA). He
provides strategic technical foreNovember/December 2004
casts for commodity and financial

markets to proprietary traders, risk
managers and brokers. E-mail:
dan@charttricks.com
References
Toby Crabel (1990), Day Trading with Short Term
Price Patterns & Opening Range Breakout. Traders
Press,
Inc.
William L. Jiler (1962), How Charts Can Help You
in the Stock Market. Trendline, Division of
Standard
&
Poor's
Corporation.
Daniel L. Chesler (1997), "Identifying Significant
Chart Formations", Technical Analysis of Stocks &
Commodities,
Volume
15.
William N. Goetzmann and Massimo Massa (2003),
"Index Funds and Stock Market Growth", Journal
of Business, Vol. 76, No. 1, 1-28.
Richard D. Wyckoff (1910), Studies in Tape
Reading.
Fraser
Publishing.
Richard W. Schabacker (1930), Stock Market Theory
and Practice. B. C. Forbes Publishing Co.
Acknowledgement
The author wishes to thank Yohey Arakawa,
Associate Professor of Japanese, Tokyo University
of Foreign Studies, for his help in finding the right
Japanese verb to describe this pattern.
27
Techniques
FINE-TUNING FIBONACCI
by Robert Miner
Price reversals are frequently made at Fibonacci retracements. After-the-fact examples of reversals at Fibonacci
support and resistance levels are easy to find but a valuable trading strategy identifies the targets for high-probability reversals in advance and while they are being
made. This article presents one such method for predicting reversals.
Key retracements
The key Fibonacci ratios for retracements
are .382, .50, .618 and .786. The last ratio,
.786, is not a common retracement but it
is a very important one. 0.786 is the
square root of .618 and is a key retracement ratio in the markets for stock
indices, bonds and currencies. 0.382 is the
least important in the series. Minor highs
and lows are made at .382 but a correction is rarely complete at or near this
retracement, and usually extends to the
.50 level or beyond. The practical trading
question is - at which retracement is a
correction likely to be complete? There
must be a filter to help identify if a reversal is probable when a retracement is
reached. Arguably, the two most important filters are the pattern position based
on Elliott wave theory and an indicator
position. The examples that follow look
at how the indicator position can be used
to identify if a reversal is probable at key
retracements.
Using indicators with Fibonacci

Figure 1 is the daily closing data of
GBP/USD from February through to
early September 2004. From the May low,
all five swing highs and lows labeled "A"
through "E" were made right on or within a couple ticks of one of the key
28
Fibonacci retracements. "A" was made at

the 61.8% retracement, "B" at the 50%,
"C" just below the 78.6%, "D" at the
78.6% and "E" at the 50%. How could
we have known which of the retracements was likely to result in a reversal?
Figure 1.
Figure 2 shows the data just beyond the

"A" high. All four retracements are
shown along with the DTosc which is a
combination of stochastics and RSI with
an adaptive 'look back' period based on
recent volatility. Just about any indicator
Techniques
may be used for this strategy including

the popular ones such as a stochastic or
RSI by themselves.
We only consider a reversal probable at
a retracement if the indicator makes a
bearish or bullish reversal at the same
time. A bearish reversal is when the fast
line crosses below the slow line and a
bullish reversal is when the fast line crosses above the slow line. The reversals do
not have to be made from the overbought (OB) or oversold (OS) zones but
if both lines of the indicator have
reached one of these zones, the trend
maybe nearly complete and about to
reverse.
The vertical line in the chart shows the
day when the bearish reversal was made
in early June. When price reached the
38.2% retracement, the indicator was still
bullish and not in the OB zone. A reversal was not likely at this retracement level.
When price reached the 50% retracement, the indicator was still bullish but
had reached the OB zone. A reversal at
Figure 2.
the 50% retracement had not been signalled but we were warned the upside
should be limited with the indicator in the
OB zone. The indicator made the bearish
reversal with price just below the 61.8%
retracement. The indicator's bearish
reversal with price at a key retracement
signalled a price top was probably complete. At the very least, the upside should
be very limited before a decline began.
Figure 3 shows the decline off the "A"
high. The indicator quickly reached the
OS zone when price reached the 38.2%
retracement, giving a warning that the
downside should be limited. The indicator's bullish reversal was made with prices
just below the 38.2% retracement, signalling a price reversal was near and the
downside should be very limited if the
pound continued to decline. The final
low was made at the next retracement 50%. Price and indicator reversals will
not always coincide exactly but a continued rally or decline should be very limited once they do. The reversal is usually
limited to the next retracement zone, in
this case, the 50% retracement.
Figure 4 shows the rally off the "B" low
and the 78.6% retracement. The chart
also shows two lines that fall just above
and below this retracement. They are the
127% and 162% external retrace-
Figure 3.
29
Techniques
Figure 4.
ments. External retracements are those

greater than 100%. In this case, they are
the external retracements of the "B"
decline. These two external retracements
are typical targets for a new high or low
when they coincide closely with
a larger retracement.
In this case, the two external
retracements of "B" and the
78.6% retracement formed a
high-probability target zone for
the next high. A reversal was
very likely at or near the 78.6%
retracement which fell near the
mid-point of the target zone.
The first close that reached the
target zone coincided with the
indicator's bearish reversal.
Prices went a bit higher the next
day making the final top just a
few ticks below the 78.6%
retracement before declining sharply.
Once again, the retracement and indicator reversal provided the high-probability
set-up for a reversal.
Of course, reversals are frequently
made at Fibonacci retracements but we
need a way to filter the market position
for a potential reversal when it reaches
each of the retracement levels. One filter
is with the position of an indicator. We

only consider a potential reversal at the
retracement if the indicator is in the OB
or OS zones or has made a bearish or
bullish reversal.
time frame for the entry trigger would be

on the hourly data.
The set-up conditions for the trade
(price at a retracement and indicator position) and entry strategy (trailing entry) are
simple and objective and meet
the two essential criteria for trading - to identify conditions with a
high probability outcome and
entry strategies with minimal
capital exposure. A tradable
reversal will not be made at every
retracement/indicator set-up.
Indicators often oscillate near
extremes in strong trends providing false reversal signals. But
when we combine Fibonacci
retracements with indicator
reversals the initial capital exposure should be minimal when the
outcome is not as anticipated.
RETRACEMENT AND
INDICATOR POSITIONS
MEET THE TWO
ESSENTIAL CONDITIONS
FOR TRADING.
30
Entries and exits

The best trade entry strategy once the
set-up conditions are in place is to trail an
entry stop one tick above or below the
prior bar, or prior minor swing high or
low, on the next smaller time frame. Since
the conditions were identified with daily
data in these examples, the next smaller
Robert Miner is manager of

www.dynamictraders.com and is the
author of Dynamic Trading published by The Traders Press.
Interview
THE TECHNICAL ANALYST TALKS TO

Jorge Bolvar
Jorge Bolvar is chairman of the Spanish Association of
Technical Analysts (AEAT), recent hosts of the IFTA
Conference 2004. He is also co-founder of Expert Timing
Systems (ETS), an advisory service for asset management
institutions. ETS was the first company to set-up an automatic link to the Madrid Stock Exchange, allowing external systematic analysis of the exchange's database.
TTA: Having successfully hosted the IFTA Conference this
year, has the conference changed your opinion of technical
analysis in any way?
JB: It hasn't led me to change any of my broad views, but it
was interesting to see how diverse the subject is in methodological terms.
TTA:So what do you think are the most exciting areas of
development in technical analysis?
JB: There are numerous good TA techniques, such as RSI,
Bollinger Bands, moving averages and so on. However, I
believe TA is limited by its own boundaries and that many
of the techniques are essentially doing the same thing. We
need to jump over the wall so to speak and explore other
opportunities around us.
TTA: So is the portfolio advice that ETS provides to asset

management companies essentially quantitative?
JB: Yes. We provide customers such as Bankinter and
Mutuactivos with asset allocation models, which they follow.
Our models are primarily based on regression analysis combined with Markowitz-based portfolio optimization. We do
however use technical trading systems for our currency
overlay management program (COM), which is a currency
hedging model designed for corporates and asset managers.
The power of COM is that we don't rely on just one or two
techniques. Instead, it combines 10 different trading strategies, including one based on mean reversion (RSI), one
based on trend following (regression) and another that
looks at the position of opening and closing prices in relation to the highs and lows.
TTA: Do you look at the fundamentals at all?
TTA: What might these be?

JB: I think TA needs to become more quantitative. I view
quants as a branch of technical analysis. We are all working
with the same raw material - price and volume - and are
working to similar objectives. The divide between traditional
charting and quants is pretty illusory. Moving averages,
Bollinger Bands, RSI are all quantitatively based.
TTA: But if RSI and Bollinger Bands are simply ways of
visualizing quantitative relationships, why not continue in
this vein?
JB: Other aspects of technical analysis are not as objective.
Patterns such as head-and-shoulders are extremely subjective and if TA stays a subjective art, it will never evolve
within the professional world. This is especially true for
portfolio management where there may be thousands of
assets to consider and where decisions need to be founded
on a systematic and well-thought out methodology. If, on
the other hand, you are trading speculatively with one or
two instruments, then speed may be more important. In
such cases, traditional charting remains extremely useful.
JB: I like to have a regular technical macro-view, i.e. technical analysis of the main macroeconomic series. This type
of analysis is very useful for gaining a global view of the
markets and the main factors affecting them
TTA: Does that mean you apply quantitative TA methods
to economic data series?
JB: Yes. There's no reason why technical analysis can't be
applied to things like inflation and unemployment data. It
provides you with a far more objective approach for gaining
a broad economic view. We use a matrix to monitor the
main trends in each of the major fundamental data series.
This is nothing new. Quantitative analysis has been looking
for relationships between economic data for years.
For more information on the AEAT, see www.aeatonline.com For more information on Expert Timing
Systems, see www.ets.es Jorge Bolvar is also CEO and
founder of TechRules.com, www.techrules.com
31
Subject Matters
THE THINKING
32
Subject Matters
INVESTOR PSYCHOLOGY
BEHIND THE CHARTS

Human flaws are consistent,
predictable, and can be
ehavioural finance is the study
exploited for
of how psychology affects
profit. That's the
financial decision making in
the markets. By examining the pervaconclusion of
sive human biases that lead people to
seemingly irrational or illogical
many prominent make
decisions, the subject offers explanations for bubbles and crashes, and
researchers
accounts for why the markets are significantly more volatile than classical
in the field of
economic research expects.
One important bias relates to the
behavioural
way
people
interpret
data.
finance.
Kahnemann, Tversky and Slovic
(1982) showed that even with good

data, people do not make decisions in
a statistically optimal way. Instead
they use rules of thumb where forecasts are often made by simple
extrapolation and "mental accounting". This less than rigorous approach
allows a host of other biases to flood
the decision-making procedure, thus
leading to decisions that may defy
logic.
This isn't necessarily bad news. If
you are aware of these cognitive
errors, then it's possible to profit
from them. Proof of this can be
found in the success of two US based
asset management companies.
Fuller & Thaler Asset Management
manages more than $1.5 billion for
both pension plans and high net
worth clients. Its investment philosophy rests on the belief that "key market participants often overreact to
negative information" and that such
participants, "naively project past,
negative information into the future

and ignore signs of recovery."
LSB Asset Management manages
approximately $25 billion. Its investment philosophy is based on,
"exploiting the judgmental biases and
behavioural weaknesses that influence the decisions of many investors.
These include: the tendency to
extrapolate the past too far into the
future, to wrongly equate a good
company with a good investment
irrespective of price, to ignore statistical evidence and to develop a
"mindset" about a company."
Some of the findings of behavioural finance run counter to technical
analysis. Take the "illusion of control". This is a bias that says humans
will see order and patterns where
none exist, a criticism that could easily be levelled at technical analysts.
Taken a step further, however, and
most of the subject's findings can be
used to explain why technical analysis
works. According to Roy Batchelor of
Cass Business School, "Behavioural
finance is finance theory that starts
with more realistic assumptions about
how people act ... The fact that people act in a certain way means that
patterns are liable to be injected into
the time series of prices."
In this article, we look at the most
important psychological biases identified by behavioural finance and suggest some general ways in which they
may relate to the observations of
technical analysis.
33
Subject Matters
Overconfidence
Overconfidence
is a tendency to
overestimate
the precision of
your information. In practical terms, this
means that you believe your own
assessment is better than others
and that your view is based on
good rational reasoning, while
others is not. Research shows this
is just as prevalent for professionals as for private investors, but is
more commonly found in men
than women (Biais et al, 2004).
Overconfidence may also arise
from previous accidental success
or the fact that investors or analysts may have some knowledge
Loss aversion
People tend to feel sorrow and
grief after having made an error
of judgement. The overwhelming
desire not to regret something is
very powerful in humans.
According to Gerald Butrimovitz,
a behavioural finance expert, the
pain of loss is 2.5 times stronger
than the joy of gains. This means
that investors and traders will
often clutch at straws to avoid a
loss, even if it means taking even
greater risks. Related to the fear
of regret, is the fear of looking
stupid. According to Rick
in a particular area.
Overconfidence isn't easily
knocked - a correct forecast will
be down to your own skill, whereas a wrong forecast is normally
due to some unexpected event
that could not have been foreseen.
On the charts?
Trends and reversals: When investors
receive news (e.g. corporate earnings),
they tend not to update their beliefs
because of overconfidence in their own
view. This behaviour gives rise to the
underreaction of prices to corporate
announcements and can lead to the continuation of existing trends when fundamentals suggest a consolidation or reversal is otherwise required. Conversely,
Harbaugh of Indiana University,

"losing any gamble, even a
friendly bet with little or no
money at stake, reflects poorly on
the decision maker's skill even
when the decision involves an
element of chance." The loss
aversion bias is commonly associated with Prospect Theory (see
box).
On the charts?
Herding: Fear of regret and fear of looking stupid are the main reasons why
people tend to go along with the crowd.
Then, if they're wrong, they can take
some consolation in not being the only
Selective memory
34
one. This leads to

continuation of
trends beyond "reasonable" levels.
Asymmetric gradient:
Loss aversion may
explain why downward corrections
tend to be steeper than upward moves,
because investors tend to hold on to
their losing positions for longer in the
hope of avoiding the pain and regret of
having made a bad investment.
Put:call ratios: Participants may take out
insurance against unlikely losses.. This
may provide justification for using the
put:call ratio as a contrary indicator for
where the market is heading.
Confirmation bias
It's not just in the financial worlds, but human development requires an ability to forget terrible times so
we can continue to function. In financial terms, this
may explain why humans fail to learn from past mistakes. We convince ourselves that the future must be
judged anew and that our current situation is special.
The tech bubble is a case in point: too many believed
in a "new economy" to justify stock overvaluations.
On the charts?
Repetitions of patterns and cycles
when investors are repeatedly hit with

similar news (e.g. good earnings surprises) they attach themselves to a new view
that is consistent with the pattern of
news. This gives rise to overreaction and
a sharp reversal.
Once a decision or opinion has been formed, individuals are inclined to pounce on information that supports their view, while ignoring or criticising information that is contrary to it. Familiar to traders as "talking your book".
On the charts?
Trend continuation: If an opinion is gaining general acceptance,
then it will be very difficult to get a contrary opinion accepted.
Thus the trend continues until the weight of contrary information is too overwhelming to ignore.
Subject Matters
Prospect Theory & Framing
Gamblers' fallacy
The gamblers' fallacy is a belief
that a trend must reverse, e.g. if
the roulette wheel has been black
for so many goes, it must be red
next. It is essentially a misguided
belief in reversion to the mean.
On the charts?
Reversals and corrections: Reversals may
only be lesser-scale corrections due to
profit-taking or covering of losses, but it
may be that humans have an inbuilt
clock or sense of geometry that says
when something just can't go on. And
these points may be described or
explained by the likes of Gann,
Fibonacci and Elliott Wave or oscillators
such as RSI and stochastics.
Anchoring
In many situations, people make
estimates by starting from an initial value or a first source of
information, which they then
adjust to reach a final answer.
Subsequent adjustment of this
assessment, however, is regularly
cut short and incomplete, resulting in too much weight being
given to the initial value. One
simple example of anchoring is
that most people think A is
People's attitude towards risks concerning gains may be quite different

from their attitude towards risk concerning losses. This was the conclusion of Nobel Prize winning Daniel
Kahneman and Amos Tversky in their
1979 work "Prospect theory: An
analysis of decision making under
risk."
In one of their experiments,
Kahneman and Tversky presented a
group of people with the following
problem:
1. In addition to whatever you own,
you have been given $1,000. You
are now asked to choose between:
A. A certain gain of $500
B. A 50% chance to gain $1,000
xx and a 50% chance to gain
xx nothing
Another group of people was presented with another problem:
2. In addition to whatever you own,
you have been given $2,000. You
are now asked to choose between:
greater than B when presented

with the following calculation
(Butrimovitz, 2004):
A: 8 x 7 x 6 x 5 x 4 x 3 x 2 x 1 = ?
B: 1 x 2 x 3 x 4 x 5 x 6 x 7 x 8 = ?
In financial terms, anchoring
means an over-reliance on a particular price level - either the
price at which the security was
bought/sold or a significant
high/low.
A) A certain loss of $500

B) A 50% chance to lose $1,000
xx and a 50% chance to lose
xx nothing
In the first group 84% chose A. In the
second group 69% chose B. The two
problems are identical in terms of net
cash to the individual; however the
phrasing of the question causes the
problems to be interpreted differently.
The way in which a question is presented ("framing") will influence an
individual's choice. Faced with a sure
gain, most investors are risk-averse,
but faced with a sure loss, investors
become risk-takers. Thus, in the financial markets an individual is likely to
frame the question in terms that
reflect his current trading position. If
they are facing a loss, they will frame
the question in a similar way to proposition 2 above. Whereas, facing a gain,
they are more likely to think in terms
of proposition 1.
Source: investorhome.com
els, because they represent prices at

which a large number of market participants sold or bought the asset.
Underreaction: Investors are anchored
to their views and are slow to
reassess them in the light of contrary news. This leads to trend continuation.
On the charts?
Support & resistance levels: Anchoring
leads to participants placing orders
or option strikes at recent or significant highs and lows, leading to the
construction of support and resistance levels. The anchoring bias also
suggests that recent very high volume periods mark out important levNovember/December 2004
35
Subject Matters
RECOGNISING PATTERNS WITH FUZZY LOGIC

by Xu-Shen Zhou and Ming Dong
One of the challenges of technical analysis is to construct models that are as successful at
identifying chart patterns as the most experienced traders and analysts.
In the article below, Zhou and Dong present one method that may help meet this challenge. Their solution is to incorporate an element of uncertainty or fuzzy logic into computer
pattern-recognition programs. This will, so the theory goes, give systems the necessary human
reasoning that allows them to detect subtle differences in chart patterns - differences that
would otherwise have been apparent only to the expert.
n classical set theory, an element

either belongs to a certain set or it
does not. In the real world, however, such certainty is often unrealistic
because of imprecise measurements,
noise, vagueness, subjectivity and so
on. For example, the concept of "tall"
is inherently fuzzy. Any set of tall people would be subjective. Moreover,
some people might be obviously tall,
whereas others are only relatively tall.
Fuzzy set theory deals with such a situation.
A fuzzy set directly addresses such
situations by allowing membership in a
set to be a matter of degree. The
degree of membership is expressed by
a number between 0 and 1, where 0
means entirely not in the set, 1 means
completely in the set, and a number in
between means partially in the set.
Figure 1 provides an example of the
concept of "income" in fuzzy sets.
Income can belong to three fuzzy
subsets - Low, Medium and High with different membership values. For
example when = 20,000, its mem-
bership for fuzzy subset Low is Low

and its membership for fuzzy subset
Medium is Medium. The solid line in
Figure 1 shows the trapezoid membership functions for the three fuzzy subsets.
The most commonly used membership functions are the triangular membership function and the trapezoid
membership function. As shown in
Figure 2, the trapezoid membership
can be fully characterised by four
parameters l, lp, r, rp marking
the values where membership equals 0
and 1 for both the left and right hand
sides of the trapezoid.
defining the three-layer comparisons

and three weights, w1, w2, w3, in the
tree. By studying each pattern on the
node of the tree, we can analyze all
the technical pattern templates.
In this study, we focus on eight pattern templates. They are head-andshoulders (HS), inverse head-andshoulders (IHS), broadening tops
(BTOP), broadening bottoms
(BBOT), triangle tops (TTOP), triangle bottoms (TBOT), rectangle tops
(RTOP), and rectangle bottoms
(RBOT). Below is an example of how
the definitions work, e.g. for HS and
RTOP:
A head-and-shoulders pattern (HS)
is characterized by a sequence of
five consecutive local extrema,
E1,E5, such that
E1 is a maximum
E3 > E1 and E3 > E5
E1 and E5 are within 1.5 percent
of their average, and E2 and E4
are within 1.5 percent of their
average.
Automating TA
The first step in automating technical
analysis is the detection of technical
patterns. We used a sequence of five
consecutive local extrema, E1,,E5,
to describe a pattern template. In general, any pattern template with five
such extrema can be described by the
tree shown in Figure 3. We can control
the shape of the pattern template by
Trapezoid Membership Function

Membership
lp
rp
Membership Value x
Figure 2.
Figure 1.
36
Subject Matters
A rectangle tops pattern (RTOP) is

characterised by a sequence of five
consecutive local extrema,
E1,...,E5, such that
E1 is a maximum
Tops are within 0.75 percent of
their average
Bottoms are within 0.75 percent
of their average
Lowest top > highest bottom
(Editor's note: Triangle tops are commonly
known as descending or bearish triangles.
The TOP is named as such by the authors
because it indicates prices will head downwards. The same principle applies to the
other continuation patterns TBOT, RTOP
and RBOT, i.e. they are commonly known
as ascending triangles, bearish rectangles
and bullish rectangles respectively).
Fuzzification process
To model the subtle differences
between patterns within the same template, we fuzzified the crisp conditions
of each pattern template by using the
trapezoid membership function shown
in Figure 2. The parameters of trapezoid membership functions for each
condition of each pattern template are
shown in Table 1. For example, the
second row of Table 1 shows the
parameters for fuzzification of the
first condition of the HS pattern template. Here, the fuzzification is based
on x, which is a measure of how high
"the head" is above "the shoulders"
Pattern/Condition
(1)
E1 < E3 ?
w1 = E 3 / E 1
w2 = E 3 / E 5
(5)
E2 < E4 ?
(4)
E2 < E4 ?
(9)
(8)
(10)
(11)
(7)
E2 < E4 ?
(12)
(13)
(14)
(15)
Figure 3.
relative to the distance between "the

shoulders" and "the body". According
to visual observation, when x=0.1 the
head is so close to the shoulders that
the entire pattern looks nearly flat and
we set the membership value in that
case to zero. When x is above 40, the
head is very high and looks like a spike
in price instead of the normal HS pattern. So again, we set the membership
value to zero. When x is in the range
(1, 5), the head, the shoulders, and the
0.1
5
0.005[(E1+E5)/2]
0.005[(E1+E4)/2]
40
0.04[(E1+E5)/2]
0.04[(E1+E4)/2]
BTOP/BBOT
Condition 1
Condition 2
0.1
1.2
0.8
2
1.2
4
10
15
TTOP/TBOT
Condition 1
Condition 2
0.1
1.2
0.8
2
1.2
4
10
15
0.005[(E1+E3+E5)/3]
0.005[(E2+E4)/2]
0.04[(E1+E3+E5)/3]
0.04[(E2+E4)/2]
w3 = E 2 / E 4
(6)
E2 < E4 ?
rp
RTOP/RBOT
Condition 1
Condition 2
Condition 3
(3)
E3 < E5 ?
(2)
E3 < E5 ?
lp
HS/IHS
Condition 1
Condition 2
Condition 3
Pattern Template with Five Extrema, Three-Layer Comparisons, and Three Weights
Table 1.
body are well placed to provide a perfect visualisation of the HS pattern. In

these situations, we set the membership value to 1.
Although the choices of variable
and parameters are ad hoc, adjustments are made so the patterns with
more symmetrical visual shapes are
assigned higher membership values.
We then obtained the pattern fuzzification membership value by averaging over all memberships within the
pattern template. For example, if the
membership values of three conditions for the HS pattern template were
0.8, 0.6, and 1.0, the membership value
of that pattern for the HS pattern
template was (0.8 + 0.6 + 1.0)/3 =
0.8.
Testing
To test our approach, we selected data
from the Center for Research in
Security Prices, a database which contains the entire universe of CRSP-listed NYSE, Amex, and NASDAQ companies. We smoothed the data in a
method similar to that of moving
averages (using Gaussian kernel
regression to find appropriate weight
ing) and then applied our algoTHE TECHNICAL ANALYST
37
Subject Matters
CARs after Completion of Eight Patterns, 19622000

CAR
A. HS, IHS, BTOP, and BBOT
0.020
CARs after Completion of Eight Patterns for Stocks with Trading

Prices of at Least $2.00, 19622000
CAR
A. HS, IHS, BTOP, and BBOT
0.015
0.015
0.010
IHS
IHS
0.010
0.005
0.005
HS
HS
0.005
BTOP
0.005
BTOP
BBOT
0.010
BBOT
0.010
0
20
40
60
80
100
120
0.015
0
Days
CAR
B. TTOP, TBOT, RTOP, and RBOT
0.04
20
40
60
80
100
120
Days
CAR
B. TTOP, TBOT, RTOP, and RBOT
0.008
0.006
RBOT
0.03
0.004
RBOT
0.002
0.02
0
0.002
0.01
0.004
TBOT
TTOP
0
0.006
TBOT
0.008
0.01
0
20
40
60
80
100
120
Da ys
Figure 4.
0.010
0
20
40
60
80
100
120
Days
Note: Asterisks indicate CARs that are statistically significantly different from 0 at the 5 percent level.
Note: Asterisks indicate CARs that are statistically significantly different from 0 at the 5 percent level.
Figure 5.
rithm to each stock in the sample for a

sample period, measuring the stock's
performance up to six months after
the occurrence of a technical pattern.
Our final sample consisted of 1,451
stocks from 1962 to 2000, for which
we detected a total of 44,150 occurrences of technical patterns (3,562 of
them belonging to more than one pattern template).
Results without fuzzy logic
We calculated cumulative abnormal
returns (CAR) for up to 120 trading
days after the completion of eight patterns. Abnormal returns (AR) for each
company on a particular day were calculated by deducting the return of a
control company for an equivalent
period. CAR is then calculated as the
sum of abnormal returns from day
38
RTOP
TTOP
RTOP
one to the day in question (range

1,,120) for each company.
CARS for up to 120 days after the
completion of eight patterns are
shown in Figure 4. The total number
of patterns detected is 44,150. Of the
eight patterns examined, CARS of
head-and-shoulders, rectangle tops,
and rectangle bottoms were statistically
significant for most of the days after
the completion of the patterns. CARS
of the inverse head-and-shoulders pattern are significant for only a few days
five months after the pattern was
detected. The magnitude of CARSs is
moderate. The highest CAR for 120
days is 3.65 percent (the CAR of
RBOT). This percentage is equal to
0.0304 percent daily abnormal returns,
on average.
Based on the definition of patterns
discussed previously, automatic pattern

detection can be greatly affected by
the prices of the stocks. For example,
a bid-ask spread of 1/16 amounts to
3.125 percent at a $2.00 price. Thus,
the stocks trading at prices below
$2.00 can easily meet the criterion of
rectangle tops and rectangle bottoms,
which require that tops and bottoms
be within 0.75 percent of their average
and the lowest top be higher than the
highest bottom. Because of this
potential problem, we removed the
patterns when stocks were trading at
prices below $2.00 on the day that the
last extreme occurred. The resulting
CARS are shown in Figure 5.
With those stocks removed, the significance of CARS for the HS and
RTOP patterns disappears. The number of days of significant CARs for
Subject Matters
We formed two sample portfolios for each

CAR
A. Portfolios Formed by Membership Value
pattern, one contain0.045
ing stocks with pattern
0.040
membership values no
larger than 0.7 and the
0.030
other containing
stocks with pattern
Value 0 to 0.7
membership values
0.020
larger than 0.7. The
results for the head0.010
and-shoulders and
inverse head-andValue 0.701 to 1.0
shoulders are shown
0
in Figures 6 and 7
0.005
respectively. Keep in
0
20
40
60
80
100
120
mind that HS are
Days
bearish technical indiFigure 6.
cators. Figure 6 indicates that the CARs
the patterns of IHS and RBOT
for the portfolio of stocks with HS
decreases dramatically. The magnitude
membership values of 0.701 to 1 were
of CARs also diminishes. The highest
mostly negative, whereas CARs for the
CAR, only 1.18 percent for 120 days,
portfolio of stocks with HS memberis less than one-third the CAR when
ship no larger than 0.7 were signifithe stocks below $2.00 were included.
cantly positive every day from the 40th
These results suggest that one needs
day to the 60th day after the 3rd day
to be cautious when using computer
of the completion of the pattern.
algorithms to detect technical patterns
Statistical tests showed that the CARs
and when interpreting the statistical
of the two portfolios were significanttests. Our results suggest that for HS,
ly different.
IHS and RBOT patterns, our fuzzy
The results for IHS shown in Figure
logic-based algorithm can be used to
7 mirror those for HS except that the
detect subtle differences even within a
CARs for the portfolio of stocks with
single pattern template.
IHS membership values from 0.701 to
1 were significantly positive, which is
Effect of pattern membership
to be welcomed given IHS is a bullish
technical indicator. We
CARs and Tests of Equality of CAR Means for Portfolios with IHS
Pattern Membership, 19622000
obtained similar results
CAR
A. Portfolios Formed by Membership Value
for the RBOT pattern.
0.02
We also formed two
Value 0.701 to 1.0
different sample port0.01
folios for each pattern,
0
one containing stocks
with pattern member0.01
ship values of 1 and
the other containing
0.02
stocks with pattern
membership values
0.03
other than 1. The
Value 0 to 0.7
results were very similar
0.04
to the two sample port0.05
folios already described,
0
20
40
60
80
100
120
i.e. we found signifiDays
cantly different postFigure 7.
CARs and Tests of Equality of CAR Means for Portfolios with HS
Pattern Membership, 19622000
pattern performances between the

portfolios for HS, IHS, and RBOT,
but not for BTOP, BBOT, TTOP,
TBOT, and RTOP.
Based on our fuzzification procedure, a pattern with a membership
value below 0.7 meets the classical definition but is visually less eye-catching.
The results suggest that for HS, HIS
and RBOT patterns, our fuzzy logicbased algorithm can be used to detect
subtle differences. To detect subtle differences in other patterns, we would
need to redefine the fuzzification procedures, choose a different x variable
for the conditions, and adjust the
parameters.
Conclusion
By introducing fuzzy logic into the
technical analysis domain, our
approach incorporates human cognitive uncertainty into automatic pattern
detection in a way that simulates
human judgement significantly better
than previously.
Our algorithm was able to detect
subtle differences for head-and-shoulders, inverse head-and-shoulders and
rectangle bottom patterns. The results
may explain why an expert technical
analyst and an average investor looking
at the same stock chart and using the
same pattern definition can derive different trading signals. Our approach,
therefore, can help investors who are
inexperienced in technical analysis to
find patterns with high membership
values and take trading positions largely (or only) in stocks with strong pattern confirmations. The result of our
CAR tests also suggest that forming
portfolios with high membership value
patterns can be profitable.
Xu-Shen Zhou is assistant professor of finance at Bloomsburg
University, Pennsylvania. Ming
Dong is assistant professor of computer science at Wayne State
University, Detroit.
Copyright 2004, CFA Institute. Reproduced and
republished from Financial Analysts Journal with
permission from CFA Institute. All Rights
Reserved.
39
Software
CBOT MARKET PROFILE
Steve Dickey explains the development and

application of Market Profile, the CBOT's
proprietary tool for displaying trading activity.
CBOT Market Profile was developed in 1982 by Peter

Steidlmayer, a Chicago Board of Trade member and Mike
Boyle, a vice-president of technology at the exchange. The
CBOT was looking for ways to give off-floor traders a better
idea of intra-day volume developments in the CBOT traded
markets. Back in 1982, the first run of clearing house volumeat-price data was not available until five hours after the market
closed. Traders that traded away from the pit could not monitor activity within the market. Consequently, it was felt that
Market Profile could be implemented as a more real-time
proxy for volume data and give off-floor traders extra information to help their trading. Steidlmayer and Boyle's intention was
to have Market Profile reflect the properties of the bell curve,
or a normal distribution, especially its first standard deviation.
This is the vertical bell curve display that is known today as
CBOT Market Profile.
Market Profile is a vertical bell curve graphic that captures
and demonstrates the developing market dynamics of price
and volume over time. It organizes seemingly chaotic price
movement and provides the trader with a means to determine
where and how buyers and sellers are interacting within the
market. Market Profile can identify prices that are acceptable
and unacceptable to the buyers and sellers.
The function of the market is to facilitate trade. To accomplish this, the market probes high prices to attract sellers and
low prices to attract buyers. An imbalance of buyers will drive
prices higher until sellers are attracted to the higher prices. The
sellers enter the market, compete, and move prices lower. This
imbalance of sellers drives prices lower until buyers are again
attracted to lower prices. Buyers enter and compete, moving
prices higher again. The result of this rotation process is the
discovery of prices that are acceptable to both the buyers and
the sellers. Market Profile organizes this ebb and flow of buying and selling by price and time to create profiles of market
activity. The typical bell curve profile is seen in Figure 1.
A vertical price scale is used which runs from the low price
up to the high price. 30 minute time increments (time brackets)
40
Figure 1.
Software
are represented by a specific letter in the alphabet. For instance

8 am to 8:30 am is represented with an upper case "A". "B"
represents 8:30 am to 9:00 am, "C" is 9 am to 9:30 am and so
on. When the market trades at a certain price in the 8 am to
8:30 am time bracket an "A" is placed in the graphic at the
appropriate price level. This continues until 9 am and then
"Bs" are placed next to the price traded and then so on
throughout the trading day. As the trading day continues a profile of the market develops.
Market Profile quickly gives the trader a view of how the
market is developing, what prices are being accepted by the
market (value area) and what prices are being rejected
(extremes). The value area is the price range traded in the day
where the majority of traders felt the price was in a fair value
Figure 2.
Figure 3.
area to trade. It is more strictly defined as the price range

where approximately 70% of total trading volume occurred the first standard deviation. The extremes, the high and low of
the day, are where few people were able to make a trade. As
the Market Profile bell curve evolves, its shape varies according
to the degree of balance and imbalance between buyers and
sellers. These varied shapes or structures are categorized as
Non-Trend, Normal, Normal Variation, Trend and Neutral
profiles. Each structure is based off of the first hour's price
range known as the Initial Balance. Another condition that
describes structure is Range Extension. This is when the price
traded moves above or below the initial balance range. (See
Figure 1)
As the Market Profile develops throughout the trading day a
Figure 4.
Figure 5.
Figure 6.
41
Software
Market Profile Structures:

Non-Trend Day
This structure has a small initial balance that contains the entire day's price range. A non-trend day
typically occurs when the market is resting or stagnating. (Figure 2)
Normal Day
The normal day has a wide initial balance price range that equals approximately 80% of the day's
total range. This is the most common pattern. (Figure 3)
Normal Variation
This structure's initial balance equals approximately 50% of the days total price range. Normal variation days have initial balance range extension in one direction or the other. (Figure 4)
Trend Day
The trend day has a small initial balance representing less than 25% of the entire day's range. This
profile shows domination by either buyers or sellers with long range extension in one direction. The
trend day will close within 5 to 10% of its daily high or low price. (Figure 5)
Neutral Day
This profile is not characterized by its initial balance but by a range extension in both directions and
a close near the center of the day's range. This structure demonstrates lack of buyer and seller conviction. (Figure 6)
certain profile characteristic forms. These characteristics or

structures can give traders clues as to the buyers and sellers'
behaviour for the day.
Market Profile not only provides information on the balance
or imbalance of current market conditions but allows each
trader to initiate positions at an individual comfort level. An
aggressive trader may
initiate a position on
the opening and use
Market Profile to monitor the trade. A more
conservative trader may
wait until a range extension has occurred
before deciding when
or where to enter the
market. Market Profile
is not a trading system
but rather an application tool that is used to
support a trader's decision making. It can be
applied to all markets traded on the CBOT from FX to commodities although most traders do not use Market Profile in
isolation, but rather in conjunction with other indicators and
techniques. Furthermore, Market Profile is not just confined to
intra-day trading but can be applied to short and long-term
analysis. Whereas conventional charting analysis is typically
derived from specific price points, Market Profile is a representation of the market as a whole and better characterizes the
actions of buyers and sellers.

Market movement is not the result of prices moving randomly; rather, it is the collective activity of market participants.
Market Profile organizes this activity into a visual structure
that defines value and allows the trader to take advantage of
opportunities where price is away from value. The majority of
users of Market Profile
are institutional traders
in the US although the
number of European
users has mushroomed
in the last few years.
This may be because of
the growth of electronic trading arcades and
the emergence of the
tool in technical analysis training courses.
Market Profile is available from CQG,
eSignal, Reuters and
FutureSource.
MARKET PROFILE QUICKLY GIVES

THE TRADER A VIEW OF HOW THE
MARKET IS DEVELOPING, WHAT
PRICES ARE BEING ACCEPTED BY
THE MARKET AND WHAT PRICES
ARE BEING REJECTED.
42
This article was prepared by Steve Dickey, vice-president

of market data products and information at the CBOT.
Additional content was contributed by Daniel M. Gramza
of Gramza Capital Management. www.cbot.com/marketprofile.
Book Review
NEW MARKET MAVERICKS
New Market Mavericks

By Geoff Cutmore
Published by Wiley Trading
229 pages, 24.95
ISBN 0-470-87046-X
New Market Mavericks is available from

The Technical Analyst bookshop at the
reduced price of 21.21 plus P+P. To
order please call 01730 233870 and
quote "The Technical Analyst magazine".
Books are usually posted within one
working day of your order.
Despite his journalistic background (as host of CNBC Europe's business breakfast
programme, Squawkbox), Geoff Cutmore writes about investing as an enthusiastic
observer rather than a hard-headed investigator. His book revolves around interviews with eight market "mavericks", all of whom share certain attitudes towards
trading and investing: Namely, recognition that buy-and-hold strategies have not
been suitable for the bear market of the last four years, nor will they be in the
future, and the belief that absolute returns are far more important than relative
ones. In essence, the book concentrates on those people that lie behind the growth
of two investment types - hedge funds and private traders.
The reader gets a real sense that the thoughts of the "maverick" are being faithfully recorded on the page. There is little attempt by Cutmore to impose himself
on his subject and he wastes little time on verifying or examining what he has been
told. Instead, what we have is eight accurate portraits written in the author's highly
readable style. The list of individuals does appear slightly arbitrary but this accusation could probably be levelled at any list designed to cover a broad range of
investment techniques. It's also the case that these are the people that have come
to Cutmore's attention while anchorman at CNBC and ones that "enjoy a loyal
public following for their insights."
The chapter on Hugh Hendry of Odey Asset Management is perhaps the most
interesting. Mr Hendry presents a very clear explanation of his investment principles (buying special situation deep value stocks that have survived the threat of
bankruptcy) and he provides practical details on how he decides on a trade (including some amount of technical analysis). Interestingly, his investment style appears
to be founded on the principle that people tend to over-react to a history of bad
news and under-react to the release of new good information, something that is
now recognised in the field of behavioural finance.
Cutmore generally covers the same topics for each individual (a biography,
investment style, trade decision-making, market outlook), but part of the appeal of
the book is that its structure is fluid enough to accommodate very different investment approaches - from Michael Browne of Sofaer Capital's use of traditional valuation techniques to Chris Locke of Oystercatcher Management's use of the astrologically derived Spiral Calendar to forecast market reversals.
For all of his subjects, you get a very clear impression of the person he is portraying, even down to height, hair colour, demeanour and working environment.
He describes Richard Cunningham of Cunningham Asset Management as looking,
"strikingly younger than his early thirties would suggest. He has a boyish oval face
topped with a neat cut of light brown hair. The sides are short over the ears and
the front combed back away from the forehead. It is a conservative straightforward look."
Many readers may find this sort of information superfluous, but there's no
doubt it makes the subject more human and you are more open to their ideas and
on guard for their fallibilities. It's a much better read for it.
But who is the book for? Interesting as it is, it's probably not riveting enough to
read cover to cover. Nor will it provide any real insight into how to trade or invest
with any particular technique. Where it comes into its own is as a broad sweep of
investment techniques. For those new to the markets or for those wishing to stepback from their normal investment approach, it may provide some inspiration. Of
course, you should also expect all the normal clichs about the psychology and discipline of trading and how to manage your money and your risk - the less experienced you are in trading, the more interesting you will find this.
43
Commitments of Traders Report
COMMITMENTS OF TRADERS REPORT

20 July - 2 November 2004
Futures only (open interest)
Non-commercial net long positions and spot rates
10-year US Treasury
Source: CBOT
10yr Treasury
5-year US Treasury
Source: CBOT
Spot
5yr Treasury
-250000
4.80
Spot
300000
4.00
4.60
-200000
3.80
250000
4.40
3.60
-150000
4.20
200000
3.40
-100000
4.00
150000
-50000
3.20
3.80
3.00
0
3.60
100000
2.80
50000
3.40
50000
100000
2.60
3.20
150000
3.00
Jul-20
Aug-03
Aug-17
Aug-31
Sep-14
Sep-28
Oct-12
Dow Jones Industrial Average

DJIA
2.40
Jul-20
Oct-26
Source: CBOT
Aug-03
Aug-17
Aug-31
Sep-14
10800
-5000
Oct-26
Source: CME
Swiss franc
11000
Oct-12
Swiss franc
Spot
-6000
Sep-28
Spot
35000
1.3
25000
1.28
15000
1.26
5000
1.24
-5000
1.22
-15000
1.2
10600
-4000
10400
-3000
10200
-2000
10000
-1000
9800
9600
Jul-20
Aug-03
Aug-17
Aug-31
Sep-14
Sep-28
Oct-12
Pound sterling
1.18
Jul-20
Source: CME
Pound sterling
-25000
Oct-26
Aug-03
Aug-17
Aug-31
Sep-14
30000
1.86
25000
1.84
Oct-26
Source: CME
Yen
1.88
Oct-12
Yen
Spot
35000
Sep-28
Spot
40000
112
111
30000
110
20000
20000
109
1.82
10000
15000
1.80
10000
1.78
5000
1.76
108
107
106
-10000
105
0
1.74
Jul-20
Aug-03
Aug-17
Aug-31
Sep-14
Sep-28
Oct-12
Oct-26
-20000
104
1.72
-5000
-30000
-10000
44
1.70
103
Jul-20
Aug-03
Aug-17
Aug-31
Sep-14
Sep-28
Oct-12
Oct-26
Commitments of Traders Report
Euro
Source: CME
Euro
3-month eurodollar
Spot
Source: CME
3-mth eurodollar
60000
1.30
Spot
400000
2.50
350000
50000
1.28
300000
2.00
250000
40000
1.26
30000
1.24
20000
1.22
200000
1.50
150000
100000
1.00
50000
0
10000
0.50
1.20
-50000
1.18
Jul-20
Aug-03
Aug-17
Aug-31
Sep-14
Sep-28
Oct-12
Nasdaq
0.00
Jul-20
Source: CME
Nasdaq
-100000
Oct-26
Aug-03
Aug-17
Aug-31
Sep-14
Oct-26
Source: CME
Nikkei
2000
Oct-12
Nikkei
Spot
-15000
Sep-28
Spot
11400
4500
11300
4000
1950
11200
-10000
3500
11100
1900
3000
11000
-5000
2500
10900
2000
10800
1850
0
10700
1800
1500
10600
1000
5000
10500
1750
500
10000
1700
Jul-20
Aug-03
Aug-17
Aug-31
Sep-14
Sep-28
Oct-12
Gold
Oct-26
10300
Jul-20
Source: CEI
Gold
10400
0
Aug-03
Aug-17
Aug-31
Sep-14
Sep-28
Source: NYCE
US dollar index
430
120000
420
Oct-26
US dollar index
Spot
140000
Oct-12
Spot
117
116
-2000
115
100000
-4000
410
114
80000
-6000
400
60000
113
-8000
112
390
40000
-10000
111
380
20000
370
Jul-20
Aug-03
Aug-17
Aug-31
Sep-14
Sep-28
Oct-12
Oct-26
-12000
110
-14000
109
Jul-20
Aug-03
Aug-17
Aug-31
Sep-14
Sep-28
Oct-12
Oct-26
45
Long-Term Technicals
LONG-TERM TECHNICALS
Provided by Thomas Anthonj, ABN Amro, Amsterdam
EUR-USD
USD-JPY
With the break above 1.2652 three weeks ago, the market indicated
that we have already resumed the bigger up-trend. Having also broken 1.2928 the big question now is where a potential top could form.
A critical area is definitely between 1.2978 and 1.3053 but if the market also breaks above the upside towards old back-calculated tops
at 1.3350 and 1.41 60 thereafter. To really harm the positive trend
the market would need to break below the last low at 1.2631.
The latest decisive break below Fibonacci-support at 106.12 is favoring the red bear-market scenario, which targets the old lows at
103.42 and 101.25 first and most likely the projected target from the
bigger H+S pattern at 95.75. To receive a first hint that this negative
scenario is not panning out the market would have to break above
108.75 and ultimately break the row of lower tops at 111.73.
GBP-USD
T-Bonds
Starting to penetrate triangle resistance at 1.8520 the market indicates that it is either forming the 3rd and completing leg up of a
countertrend rally to 1.8748/71 (76.4 %/last top) or, once the latter
would give way, has already resumed the bigger up-trend with very
bullish implications medium.-to long-term (minimum target 2.0115).
To reverse this positive picture the market would need to break keysupport at 1.8196/63 and ultimately triangle support at 1.7800 targeting 1.7375 thereafter.
Bouncing off from the weekly trend channel support (now at 103-19)
the market is most likely performing another internal 2nd wave
rebound that is expected to stall somewhere below 116-12. If 110.21
(last low) is broken we would be fairly sure the bigger bear-trend has
resumed.
46
Long-Term Technicals
Dow Jones
S&P 500
The correction pattern shown from the 3rd wave top at 10747 is
clearly looking like a completed double zigzag now although falling
slightly short of the projected target at 9479. It would now only take a
decisive break and close above 10507/10571 (76.4 %/old top) to
receive the final evidence that the up-trend has resumed targeting
old tops at 11350 and 11750 from where a bigger setback is expected to unfold again.
The break above key-resistance at 1139/46 has clearly indicated that

the correction is over. But looking at the picture above it is pretty
obvious that we are in a completing 5th wave advance that could
start a nasty correction (61.8 % or 76.4 % of the advance from 768)
in case the market fails to clear strong resistance at 1175/90. The
next critical area above would then be around 1246/53. A break
below 1142 would now be the first sign of weakness whereas it
would require a break below the last low at 1090 to call for a deeper
setback to at least 1061.
Nasdaq
Nikkei
Bouncing off from trend line support at 1800 the market would only
need to settle above key-resistance at 2056/59 (old top/76.4 %) in
order to confirm the resumption of the bull-trend. However, as this
advance could easily complete the first 5-wave structure up we
strongly recommend taking profit towards the old 2328 top as the
market might again run into a bigger setback there under that could
potentially retrace 61.8 % or 76.4 % of the advance from 1108.
The choppy sideways trading action of the last weeks and months
doesn't deliver any hints yet whether we are finally out of the doldrums. The market would need to break the row of lower tops at
11410 in order to at least test key-resistance at 11789/11969 (76.4
%/old top). Only a break above the latter would eliminate the threat
of missing the left shoulder of a bigger inverted H+S pattern down to
9383. A break below trend line support at 10601 would point in that
direction.
47
Training and Events Diary
TRAINING AND EVENTS DIARY
NOVEMBER DECEMBER JAN - MAR
25/26
11-22
Course:
An introduction to charting &
technical analysis
Organiser:
International Petroleum Exchange
Contact:
training@theipe.com
Event:
STA meeting
Organiser:
Society of Technical Analysts
Contact:
info@sta-uk.org
Course:
STA diploma course
Organiser:
Contact:
info@sta-uk.org
JANUARY
FEBRUARY
MARCH
12
28
17/18
Event:
STA meeting
Organiser:
Contact:
info@sta-uk.org
Course:
Introduction to technical analysis
Organiser:
Quorum Training
Contact:
course@quorumtraining.co.uK
SUBMISSIONS FOR
EVENTS & COURSES
IN 2005
Please email us at:
editor@technicalanalyst.co.uk
48
Course:
Technical analysis and charting
Organiser:
Chartwatch
Contact:
chartwatch@aol.com

Technical Analyst

Transféré par

Informations du document

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

Technical Analyst

Transféré par

Droits d'auteur :

Formats disponibles

nov/dec 2004

The publication for trading and investment professionals

The January Barometer

London Stock Exchange Advertisement

CONTENTS 1 > FEATURES

GET QUALIFIED IN TECHNICAL ANALYSIS

Monthly lectures and regular teaching courses in technical analysis

The STA Journal publishes research papers on TA techniques and approaches

Provide members the opportunity to discuss technical approaches and markets

The STA Diploma Exam is internationally recognised as a professional level qualification

For more information on how to join and what is involved in passing

To subscribe to The Technical Analyst magazine

us on +44 7000 710207

I/We would like to subscribe to

CONTENTS 2 > REGULARS

Unit 201, Panther House,

THE TECHNICAL ANALYST TALKS TO

For information, please contact:

COMMITMENTS OF TRADERS REPORT

THE TECHNICAL ANALYST

eSIGNAL ADDS BRAZIL AND

to find market-moving news and we

Eurex has reported record

UPDATA RELEASES MARKETHAWK

Charting software provider, Updata,

THE TECHNICAL ANALYST

charts, MarketHawk can also send

BIS 2004 REPORT:

The Bank of International Settlements (BIS)

Foreign exchange average daily turnover ($bn)

Turnover share by currency pair (%)

THE TECHNICAL ANALYST

IFTA CONFERENCE IN BRIEF

Marc Michiels with conference speakers

The annual IFTA conference held in Madrid in November gave market

RSI TOPS POLL

Independent analyst Gerald

TA guru, John Murphy's talk

THE TECHNICAL ANALYST

After that, its downhill for a

Whilst DeMark is only around

John Bollinger, not surprisingly, stuck to Bollinger

Manfred Huebner of Deka

today. Bollinger also touched

most recent price action. If

achieved winning trades of

THE TECHNICAL ANALYST

y early November, USD/CAD had depreciated by

and slow stochastic studies reached 13-year oversold

THE TECHNICAL ANALYST

Prices reaffirm long-term downtrend

WHILE THE LONG-TERM OUTLOOK IS

Retracements to provide selling opportunity

re-enter the channel and confirm the

George Davis is chief technical analyst with Royal

THE TECHNICAL ANALYST

ELLIOTT WAVE OUTLOOK FOR

he long-term outlook for gold continues to suggest

THE TECHNICAL ANALYST

ULTIMATELY A 5 WAVE ADVANCE WILL

appear to be consolidating their breakout above $430, and

advance from 35.30 is now complete and that the index is

THE TECHNICAL ANALYST

OUTLOOK FOR USD/CHF