Vous êtes sur la page 1sur 20

The STA is delighted to be hosting the IFTA conference in London this

month. It is a measure of what an important event this has come to be


for technical analysts around the world that we have managed to attract
such a galaxy of international speakers all of whom are distinguished in
their particular branch of analysis. The conference could not be taking
place against a more challenging market background (the old Chinese
curse May you live in interesting timessprings to mind) and we look
forward to hearing what each speaker has to say. For those of you
unable to attend, there will be feedback in the next issue.
The bear market does seem to have brought with it some very
favourable press coverage for technical analysis. In August, for example,
the US magazine Barrons posed the question would technical analysis
have told us to get out of stocks like Enron or WorldCom?to which the
author, Michael Kahn, came to an encouragingly positive conclusion.
One of the objectives of the Market Technician is to provide a sounding
board for new work and ideas that members are in the process of
developing as well as a forum for discussing why certain types of
analysis do not work in some circumstances. A surprising failure earlier
this year was the Coppock, one of the most well-established and
hitherto reliable long term indicators. Its upturn earlier in the year
prompted considerable interest in the equity markets. In the event, the
buy signal proved to be premature. Investment Research of Cambridge
has been keeping records of this indicator for the UK market since 1965
and this is the first time that they are aware of it producing an
inaccurate call on the market. The records for the US market go back to
1901 and of the 30 buy signals that it has generated only three have
failed. These were in 1901, 1914 and 194; the last two were notably
times of world war. Perhaps the gathering storm clouds over Iraq were
responsible for this latest failure?
The Societys Diploma Course on technical analysis at South Bank
University attracted a record number of students this year and most of
them went on to take the exam afterwards. It is pleasing to report that
the results showed a marked improvement on previous years, with the
number of both passes and distinctions having risen proportionally.
This improvement is even more impressive given the fact that the pass
mark was raised in 2001. Under such circumstances there is a natural
tendency to query whether the standard required has fallen.Whatever
may be happening elsewhere in the educational field, this is not the
case with our exam. It has been tightened up to such an extent that
questions have been raised about whether the standard is too high. We
do not think it is and this years results are surely proof of this.
It is hard to pin point any particular reason for such a major
improvement. One factor was clearly the Revision Day, which this year
was enlivened by a gas leak, but notwithstanding this distraction a lot of
hard work was put into the sessions and it clearly paid dividends.
Another factor was the quality of this years group of students. They
were very enthusiastic and worked extremely hard to develop their
knowledge of the subject. The successful candidates are listed on page
two congratulations to them all.
IN THIS ISSUE
STAExam Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2
M.Feeny Intellectual property rights
prising open the black box . . . . . . . . 3
D.Watts Asurvey of technical analysis
charting . . . . . . . . . . . . . . . . . . . . . . . . . 4
G.Celaya Software review . . . . . . . . . . . . . . . . . . 7
B.Jamieson Book review . . . . . . . . . . . . . . . . . . . . . . 7
A.Hill Using Candlesticks to play the
trend . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
P.Beutell Further evidence of a Great Bear
Market . . . . . . . . . . . . . . . . . . . . . . . . . 10
J.du Plessis Does continued use of the wrong
method make it right? . . . . . . . . . . . 12
T.Plummer Pattern and periodicity in financial
circles . . . . . . . . . . . . . . . . . . . . . . . . . . 14
S.Griffiths Identifying latent energy in the
markets . . . . . . . . . . . . . . . . . . . . . . . . 20
October 2002 The Journal of the STA
Issue No.45 www.sta-uk.org
MARKET TECHNICIAN
COPY DEADLINE FOR THE NEXTISSUE
January 2003
PUBLICATION OF THE NEXTISSUE
March 2003
FORYOURDIARY
10-12th October IFTAConference,London
Wed.13th November Software Solutions,
A General Survey of Technical Analysis Software
David Watts
Wed.4th December Equity Markets in 2003
Nicola Merrell,Technical Analyst,
JP Morgan Securities
N.B.The monthly meetings will take place at the
Institute of Marine Engineers
80 Coleman Street,London EC2 at 6.00 p.m.
MARKETTECHNICIAN Issue 45 October 2002 2
STADIPLOMA EXAM RESULTS
SPRING2002
DISTINCTION
Fabio Fraschetti
Walid Moukarzel
Tarlock Randhawa
PASSLIST
DITA II CANDIDATES
PASS LIST
Eric Lanteigne
Dirk Bach
Robert Freuchtl
Irfan Soezen
Cyril Marini
CHAIRMAN
Adam Sorab,
Deutsche Asset Management,
1 Appold Street,London EC2A 2UU
TREASURER
Vic Woodhouse.Tel:020-8810 4500
PROGRAMME ORGANISATION
Mark Tennyson d'Eyncourt.
Tel:020-8995 5998 (eves)
LIBRARY AND LIAISON
Michael Feeny.Tel:020-7786 1322
The Barbican library contains our collection.Michael buys new books for it
where appropriate.Any suggestions for new books should be made to him.
EDUCATION
John Cameron.Tel:01981-510210
Clive Hale.Tel:01628-471911
George Maclean.Tel:020-7312 7000
EXTERNAL RELATIONS
Axel Rudolph.Tel:020-7842 9494
IFTA
Anne Whitby.Tel:020-7636 6533
MARKETING
Simon Warren.Tel:020-7656 2212
Kevan Conlon.Tel:020-7329 6333
MEMBERSHIP
Simon Warren.Tel:020-7656 2212
Gerry Celaya.Tel:020-7730 5316
Barry Tarr.Tel:020-7522 3626
REGIONAL CHAPTERS
Robert Newgrosh.Tel:0161-428 1069
Murray Gunn.Tel:0131-245 7885
SECRETARY
Mark Tennyson dEyncourt.
Tel:020-8995 5998 (eves)
STA JOURNAL
Editor,Deborah Owen,
108 Barnsbury Road,London N1 OES
Tel:020-7278 4605
Please keep the articles coming in the success of the Journal depends
on its authors,and we would like to thank all those who have supported
us with their high standard of work.The aim is to make the Journal a
valuable showcase for membersresearch as well as to inform and
entertain readers.
The Society is not responsible for any material published in The Market
Technician and publication of any material or expression of opinions
does not necessarily imply that the Society agrees with them.The Society
is not authorised to conduct investment business and does not provide
investment advice or recommendations.
Articles are published without responsibility on the part of the Society,
the editor or authors for loss occasioned by any person acting or
refraining from action as a result of any view expressed therein.
Networking
WHO TO CONTACTON YOUR COMMITTEE
Exam results
ANYQUERIES
For any queries about joining the Society,attending one of the
STA courses on technical analysis or taking the diploma examination,
please contact:
STA Administration Services (Katie Abberton)
Dean House,Vernham Dean, Hampshire SP11 0LA
Tel:07000710207 Fax:07000710208
www.sta-uk.org
For information about advertising in the journal,please contact
Deborah Owen
108 Barnsbury Road,London N1 OES. Tel:020-7278 4605
Hakon Anderson
Karri Andrews
Derek Cilliers
Matthew Clements
Andrew Coles
Paul Denley
Christopher Farrand
Darren Hayward-Williams
David Jenni
Alison Karlin
Robin Kidman
Clive Lambert
Henry Law
James Lewis
Martin Llewellyn
Jonathon Marriott
Robert Melling
Jason Moores
Cormac OConnell
David Owen
Matt Peters
Malcolm Pryor
Ketan Raja
Richard Ramyar
Sara Raynor
David Reddy
George Roberts
Roger Sanders
George Sharpe
Toby Sheppard
Sam Stanley
Adaiklan Subramanian
Ewart Sutton
Mark Thompson
George Traore
Eoin Treacy
Richard Waldron
Henrik Wigernas
Oliver Wilson
Scott Yarwood
Andreas Papdopoulos
Matey Gerov
Peter Knowles
Georgiou Neophytos
Kyprianou Kyprianos
Issue 45 October 2002 MARKETTECHNICIAN 3
Does the purchaser of a black box trading system
acquire full rights of ownership?
According to a legal judgement by an English court,the owner of a
machine has the full right of ownership and with that goes an entitlement
to dismantle the machine to find out how it works and tell anyone the
owner pleases.
Even if the seller had expressly stated that the information was
confidential,this would not work to override the owners entitlement to
find out how the machine operates.
To me,this seems a pretty important judgement.
Lets look at the whole issue of commercially confidential information
under English law,and then specifically at the legal judgement on the
confidentiality of how a black box works.
Suppose you publish a newsletter containing proprietary trading signals
or other commercially-sensitive confidential information which you
broadcast by e-mail. What principles apply when an unintended or
unauthorised addressee receives it?
As soon as an addressee realises they are not an intended or authorised
recipient of obviously confidential information,they come under a legal
duty of confidence to stop reading the e-mail and not to use or disclose
the information. In practice the recipients curiosity is likely to get the
better of them,and in any event the damage will often have been done
by scanning the first few lines. The court would,if necessary in those
circumstances,grant an injunction to restrain the use or further disclosure
of the confidential information.
When the recipient,without turning a blind eye to suspicious
circumstances,is honestly unaware that an e-mail has been sent to him by
mistake or he honestly believes the sender had the right to disclose the
information,he is free to use and disclose the information,until such time
as he becomes aware of the breach of confidentiality. He can subsequently
come under an obligation of confidence upon discovering the truth,but he
will not be prevented from using the information where he has already
done something to make himself worse off by acting on that information,
or the information has already entered the public domain. For this reason
recipients of a leaked or wrongly addressed e-mail should be notified as
soon as possible following the discovery of the leak.
In the case of anonymous disclosure,a vast amount of news,rumour and
gossip swirls around markets,particularly the foreign exchange market,
concerning who is buying and selling what. This is price-sensitive
information relating to individual trades usually supposed to be
confidential to the counterparties and their brokers.The legitimacy of
using anonymous data was considered by the court in the Source
Infomatics case. The question was whether a company which compiled
information about doctorsprescribing habits was entitled to sell that data
(with patientsdetails suppressed) to pharmaceutical companies to help
them market their products. The UK Court of Appeal concluded that there
was very little authority on the subject,and ruled that it was necessary to
look in each case at what the law of confidentiality was concerned to
protect. In that particular case,providing patientsprivacy was not put at
risk,there had not been unauthorised use of confidential information. The
fact that the patients had not consented (and may not have consented if
asked) was not overriding.
This decision is important to parties who publish or disclose statistics as
part of the service they offer to customers,brokers,databases and on-line
exchanges. It shows the courts will take a pragmatic approach to the use
of data on an anonymous or aggregate basis where it can be
demonstrated that the interests of the party to whom an obligation of
confidence is owed would not be prejudiced by publication.
As English law requires criminal and civil trials to take place in public,any
information relied on in a trial will become public knowledge. Companies
therefore traditionally prefer to resolve commercial disputes in the privacy
of arbitration. A court did recently protect the confidentiality of a
document beyond the duration of the trial in which it featured,but only
because access was not required by those involved in the trial nor by an
interested spectator to the trial. If a dispute does end up in court,parties
should identify which documents are confidential and take steps to
exclude them from disclosure if not relevant,or blank out confidential
parts with the agreement of the other parties.
What happens when an employee leaves and goes to a
competitor?
The courts have to strike a balance between protecting the interests of
the ex-employer and allowing the individual to use his skills and
accumulated knowledge to find another job. Allegedly confidential
information would need to have special qualities for it to be classed as a
trade secret,on the facts of the case,and be restricted in dissemination
such that its disclosure would cause real or significant harm. The courts
are reluctant to extend such protection. In a recent case where a business
plan was based around a training programme,the information was not
judged to be a trade secret because the constituent parts were all
available on the internet,albeit separately.
A business with confidential information which could be taken to a
competitor should identify the specific categories and put confidentiality
clauses and enforceable restrictive covenants into the employeescontract
of employment;also look into the possibility of patent protection.
Regarding technical information,equipment that is sold often contains
proprietary technology or clues about how it is manufactured that the
seller wishes to keep confidential. For instance suppose a black box
trading system containing various indicators and oscillators and
algorithms to generate trading signals,is put in a sealed box with a sale
price of $10,000 and it does successfully trade the markets.The box
contains some commonly available technical analysis tools and relies on
well-known technical analysis techniques. Can the seller prevent the
purchaser discovering for himself how the equipment was made or put
together,by sealing the box and informing the purchaser that the
contents of the box are confidential?
This was the issue that the court had to decide in the Mars case.Mars was
a leader in the design and manufacture of mechanisms to check coins fed
into vending machines. A company bought one of the most sophisticated
of these mechanisms and reverse engineered it,breaking its operating
code which was protected by encryption. Mars claimed that the
purchasers reverse engineering amounted to a breach of confidence. But
the court rejected the claim,finding that the owner of the machine has
the full right of ownership and with that goes an entitlement to
dismantle the machine to find out how it works and tell anyone it
pleases.
Even if Mars had expressly stated the information was confidential this
would not work to override the buyers entitlement to find out how the
machine worked.
So the lesson of this case is that once a process,technique or code is
embodied in a product that is sold,it can no longer be regarded as
confidential and the buyer cannot be prevented from reverse
engineering it.
Manufacturers and developers should therefore use the other means
described above,such as confidentiality clauses with contractors and
partners,restrictive covenants in employment contracts,and arbitration as
an alternative to the court,to protect their intellectual property.
Intellectual property rights prising open
the black box
By Michael Feeny
MARKETTECHNICIAN Issue 45 October 2002 4
Introduction
Since the publication of the last software survey the number and
sophistication of technical analysis software continues to be ever growing,
such that the scope of this survey can only hope to catch a sample of
proven software and a variety of new releases.
Good news for the newer technician is that it is now possible to obtain a
competent charting package with a quote downloader for under the 50
mark. This is a remarkable development brought about partly as a result of
the market boom and competition between vendors over the Internet.
Then free quotes from the likes of Yahoo have lead to the development of
a number of downloader programs that can obtain quotes on up to 10,700
securities free of charge.While the data quality may be questionable,for
many exchanges the data is clean and perfectly acceptable for accurate
charts to be produced. These developments would have been just a dream
a few years ago,now made reality by the power of the Internet.
A price database of this magnitude would have cost many thousands of
pounds not that many years ago. A listing of both the available packages
and downloader programs are given in the tables.
For the professional the quality of the service is of prime importance,but
here again Reuters Bridge has such competitive pricing for a quality
provider that a professional System becomes increasingly available to the
stand-alone trader.
One unwelcome development is the return of TA programs tied to the
hard drive by means of an authorisation code. Hard drive crashes are still
not unknown,with the possible fatal consequence that you may loose
both your hard drive and your TA program. As it is not unknown for a
software house to go into liquidation, it is always best to request the
cumbersome Dongle if possible or buy only those programs like
Metastock,with no tied security at all,just a call to the original CD some
months after installation.
Of note this year has been the takeover of Bridge by Reuters and the very
competitive pricing of the new Bridge Internet services,with professional
level data quality at a very low price.
There are still only a comparatively small number of packages and
charting services that offer multi-market and sector analysis for the equity
professional. Of note is the introduction of Market-Master to this
country.It has superb data handling and just awaits a data service offering
world market coverage to make this an attractive package for the equity
analyst. For the more mathematically gifted,the new XperTrader will
offer spectral analysis amongst other advanced proprietary studies. While
one of my own favourites is MTpredictor for elegant wave and time price
analysis,but not least because of Steves continued support to users via
his support group and his free commentary service with suggested trade
set-ups to make this package real value.
As always the pressure to release new software after substantial
development cost is enormous and there has been a number of
complaints from members over buggy software and datafeeds. I have
always tended to take the attitude that,when buying newly released
software,there will be problems to be sorted and reported at least until
version 2 is released.Buyers should always order the demo or trial
software,to make sure they are happy with the charts or service offered.
I thought it worth highlighting some of the best from this years software
survey.
Free web services
www.Bigcharts.com and www.Stockcharts.com are the most widely used
for charts with www.advfn.com and www.FT.com both being widely used.
Web-based services
US services like www.Iqcharts.com appears highly competitive for
realtime stock market data. While www.Quote.com and www.ADVFN.com
have many fans at a slightly higher price level. Then Reuters Bridge looks
hard to beat for the coverage and range of markets offered.
eSignal has a new front end which spruces up the quote and chart pages
www.dbc.com
Software packages
At the entry level the UK Sharescope and US TC2000 by Worden
brothers both attract many advocates. A good value package is
Gannalyst Lite for $65,the chart package is all that many would ever
want and you get free quotes from Yahoo as well. (Pity about the
authorisation code protection tied to the hard drive)
System testing
New is Weathlab 2.1 for portfolio level testing and Tradestation has now
gone internet based with a brokerage for full automatic execution (if you
trust the system with your money). But note that a reasonable EOD
service is also available for $50 per month.
Professional level services
Commodity Research Bureau still exists and continues to provide those
great wall charts to impress your clients and produce those old-fashioned
chart books for the desk. Reuters Bridge looks to offer great value via the
Internet,but with the attendant risks should the Internet ever fail.
Professional Dealing Room Packages
A survey of technical analysis charting
By David Watts
Information
System
Web Address Packages
Available
Comment
Bloomberg www.bloomberg.com Various.Primarily
known for their
support to bond
traders.
Mainstream US service
provider.Popular with
Bond desks.Extensive
web services including
Bloomberg Radio and TV.
Bridge
(Online
appllication)
www.bridge.com Various.Noted for
their Worldwide
market coverage.
Now owned by
Reuters
Reuters Bridge
Channel is just $79
per month. Reuters
Platinum Personal
Bridge -$25.00 Per
month
Bridge In Touch - For
Portable Devices
$20 per month.
Now delivering state of
the art web based
services with very
competitive pricing for
the trader,analyst or
investor.
The Reuters Personal
Bridge is excellent value
for money and why not
have chart updates
straight to your PDA?
Commodity
Research Bureau.
Now owned by
Logical Systems,
Inc.of Chicago
www.crbtrader.com Known for their
commodity chart books
since 1934.Also provides
long-term wall charts.
Extensive Commodity
database available.
SystemMaker is
available for those who
wish to back test their
trading systems with
many features for the
Forex and Commodity
markets incorporated.
Commodity
Quote Graphics
www.CQG.com A good TA charting
front
end that also
interfaces a wide
variety of TA
software.
CQGNet is the web-
based service.
An established supplier
to the commodity
trading community.
Known for their timely
and clean data.
CRBcharts is a low
cost $20 per month,
end-of-day charting
service. Also
SystemMaker is a
powerful technical
analysis program
that easily backtests,
trading systems.
An extensive
collection of chart
books and
Commodity data is
available.
Datastream
Thomson
Financial.
www.icv.co.uk Datastream Advance
4 provider.
Publishes the famous
Extel surveys.
Provides Stockbroker
information via the
Datastream Service.
Previously owned the
Marketeye Private
trader service.
Issue 45 October 2002 MARKETTECHNICIAN 5
Stand Alone Technical Analysis Software PC
Technical Analysis Toolbox Software Mac.OS
Pattern Matching Software
Information
System
Web Address Packages
Available
Comment
Reuters www.reuters.com
www.reutersdatalink.com
Now with Metastock
Professional as a
front end TA
package.
Long established as a
premier provider to the
Forex Market.The US
service Reuters datalink
offers a range of North
American data packages.
Tradermade
International
Limited
0208 313 0992.
www.Tradermade.com Tradermade
Workstation &Web
are just two of the
available products.
Web package
enables easy chart
distribution..
Extensive range of
services.
Another long established
TA data provider for the
professional.
Onlineservicealso
availablewith training
and UK support.
Technical
Package
Web Address Packages Comment
AIQ Systems
001-800-332-2999
www.aiqsystems.com/
www.aiqsystems.com
/uk.htm
UK email:
sales@aiqsystems.co.
uk
TradingExpert Pro
TradingExpert EOD
Monthly plans available
from:
$59 for delayed data or
$79 for RT(+exchange
fees) both an internet
data feed that interfaces
with the Track Data
package,Mytrack.
A highly regarded
package for the Equity
Markets.Extensive sector
analysis can be
performed.With the
proprietary expert
trading system
UK support available.
Equis
International.
Tel:
001 800 882 3040
www.equis.com Metastock Professional
for eSignal $1495
MetaStock Prois also
available for Reuters.
Metastock EOD $399
MetaStock Online
Free
Downloader Free
Metastock,is a long time
favourite with technicians,
due to a robust program,
good support and the
range of TA studies.
Able to test simple
technical systems.
Metastock online for
free US equity charts.
USsupport only.
INDEXIA.
Tel.01442 878015
Now owned by
Updata
www.indexia.com INDEXIAhas numerous
services.EOD,RTand
option trading.Packages
available.
Indexia and Indexia II
Plus.
An Established UK
software house that
provides a wide range of
studies plus the excellent
propriety Indexia Filters.
Dos type interface.
With a knowledgeable
and professional UK
Support service.
ShareScope www.sharescope.
co.uk
or email
orders@sharescope.
co.uk
Membership Fee of
79.95 (inc VAT).
EODMonthly
subscription of 11.95
(inc VAT).
RT84.95 (inc.VAT) per
month
Demo available via the
web site.
A popular award winning
charting and data
package used.
Lacks some of the more
sophisticated studies and
extensive fundamental
information.
Stratagem
Software Intl
Tel:
001 (504) 885-7353
http://209.123.116.59 SMARTrader 4.01
$299.0 and ,
SMARTrader RT 4.01
$995.00
QuickCharts,is $99.00
RT package interfaces
with Quote.com
SMARTrader the
upgrade path for
Computrac users. A
Computrac for Windows.
It has all the standard
Computrac indicators
plus more.
USsupport.
AGET
Tel:330 645 0077
www.advancedget.
com/index.html
Advanced GET $2,995
Advanced GET RT
$2520 for a one year
lease
A highly regarded
package due to its ease
of use,with one click
trendlines.Provides
indicative Elliot Wave
counts and Gann Boxes.
Scanner available
USsupport only.
Tradestation
Securities
Tel.:
001 800 808 9336
Or
001 888 853 9741
www.tradestation.com Tradestation securities
onlinewith various data
packages and a tied in
brokerage.
Tradestation Real Time
and EODboth allow
system testing.
Now a web based
service with systems and
order interface.Non
brokerage fees are RT
$199.95 per month and
$49.95 EOD.
Tradestation is now
online.Still with the
original easy language
for programming
systems.
Non brokerage options
allow you to design and
test systems online with
US data supplied.
TS200i is still supported
in the UK via
www.scapler.co.uk.
Technical
Package
Web Address Packages Comment
Updata
Tel:
020 8874 4747
www.updata.co.uk Updata Technical
Analyst
Updata Trader Pro
First Year 990+VAT
(Real-Time)
Technical Analyst is a
new release. With a
number of packages at
various levels of service.
Also supports the
Fairshares TA packages.
UK support.
WaveWise
Market
Spreadsheet.
Tel:
001 908-369-7503
www.members.aol.
com/jtiware/
Waverwise costs
$299.00
A spreadsheet interface
and good charts make
this a flexible data
handling and charting
package.For the
spreadsheet lover.
Worden Brothers
TC2000
www.tc2000.com Free software but tied to
their data EOD service at
$29.75 per month.
Covers US equities.
Version 4.7 released.
Well know US software
package for tracking US
equities.Famous
propriety indicators
Balance of Power and
Time Segmented Volume.
Synergy Software
Tel:01582424282
www.synsoft.co.uk For the professional and
dealing room:
Sequencer and RITA
For the private investor:
Portfolio Evolution 169
+ VATper annum
Portfolio Advantage
395 + VATper annum
Portfolio DayTrader
(Real-time) 890+VAT
plus Exchange Fees
Plusdecoder box
155+VAT
Both investor and
professional packages.
The ease of Relative
Strength charting has
always been a strength
of the Synergy software
packages.
UK Support.
Technical
Package
Web Address Packages Comment
Behold for the
Mac
www.bhld.com Behold Version 2.9 - $795
Now able to be run on
the PC with Mac
emulation.
Behold for the Mac is still a
fine product with the
ability to test EOD systems.
Behold for the Mac .
With the original Easy
language from the early
Omega System Writer
Free update policy and
US Support.
Investor RT
Charting
Tel:
001800546-6842
or
001404733-5733
www.linnsoft.com Investor/RT$1495 or
$595 per year
Cross platform TA
package with extensive
Studies.
USsupport.
ProTA by Beesoft
(Online ordering
and support)
http://www.beesoft.
net
ProTA
$59
ProTA Gold $199
Extensive studies for a
competitive price.With a
wide range of data
formats supported
including,Metastock
USSupport.
Trendsetter
Analyst
Tel:
800-825-1852 U.S.
or
001 (714) 997-9775
www.Trendsoft.com/ Personal Analyst $259
Personal Hotline $495
Pro Analyst
$59 per month
An extensive range of
Software.The Pro-Analyst
package Includes a
day-trading Pivot based
trading system.
USSupport.
Technical
Package
Web Address Packages Comment
Dynamic Trader www.dynamictrad
ers.com
Dynamic Trader V4
$1700.00.
RTadditional
$33.00 per month.
Classic time and price
analysis
Package by Robert Miner.
Pattern Smasher http://www.
kasanjianresearch.
com
Pattern Smasher EOD
$1895.
RTprogram in
development
Pattern Smasher can test
and scan for pre-defined
bar patterns.Has a wide
range of pre-programmed
patterns such as double
tops and bottoms.
USsupport and ongoing
Newsletters.
Patterns http://www.markets
online.com/
Patterns
$429
The updated Nava-Patterns
program.Dos based.Able
to test bar by bar and
candle patterns.
USsupport.
MARKETTECHNICIAN Issue 45 October 2002 6
Portfolio Level System Testing
Internet Based Charting,Data or News Services
Gann Charting
Free Software
Just Released
Technical
Package
Web Address Packages Comment
MTPredictor http://www.mtpre
dictor.com
MTPredictor 2.0
is $1,495 (plus VAT)
Able to recognize certain
Elliot wave setups and
propriety time-price
patterns.
With extensive email
support and a trading e
group you get much more
than just the software.
UK support.
Prognosis
Software
Development
www.elwave.com
www.prognoss.nl
For those who want
automatic Elliot wave
counts generated with
Projected wave targets.
Another program to
consider is Elliott Wave
Analyzer III at :
http://www.elliott
software.com
Technical
Package
Web Address Packages Comment
Trading Recipes
Tel:410 263 0798
None Trading Recipes
$2295.00
R W Systems
Address:
5757 Westheimer,
Houston,TX 77057.
Dated but flexible system
testing package.Basic
type language.
USsupport.
TechniFilter Plus
Tel:919 856 9600
www.rtrsoftware.
com
TechniFilter Plus V.8
$425
TechniFilter is a reporting
and system-testing
package for Windows but
with only basic charting.
Excellent fast scanning.
A comprehensive,
proprietary system testing
language that takes some
time to learn but does
allow portfolio level
testing.
Web based support only.
WeathLab
Developer
V2.1
http://www.wealth
-lab.com
Weathlab 2.1 $650 Both
EOD and RT.
30 day trial available.
Weathlab combines TA
charts with the ability to
test a system across a
portfolio.WealthScript
Programming language.
Technical
Package
Web Address Packages Comment
Advanced
Financial Network
www.adfn.com Unlimited free real time
prices.
Level 2 Datais from
35+VATper month.
A rival to Quote.com for
UK equity quotes.
FT.com www.FT.com A vast site with news and
FTinvestor services.
Charts and news.The
charting service has been
reduced but still a great
site.See Bigcharts.com
for the OHLC charts.
BigCharts www.Bigcharts.com Interactive charts and
quotes.BigCharts is a
FREEservice.
A provider of charts to
many other internet
sites.Excellent UK share
charts available.
Mytrack www.mytrack.com Delayed quotes are free
for both UK and US
equities
Mytrack provides a
spreadsheet type
interface to a quote table.
Basic charts available or
it can interface to the
AIQ charting package.
NexTrend www.nextrend.com NexTrend Trader From
$15 per month RTand
Nextrend Trader Plus
From $79.95 a month
Wide range of TAstudies
and charts.
US Securities and Futures
data only.
Technical
Package
Web Address Packages Comment
Proquote.com www.proquote.com Proquote Service
Fees From 45.00 per
month (+ VAT).
Professional feeds from
100 per month.
Equity-based quotes and
charts.Primarily a service
to equity traders and
investors.
Siliconinvestor
IQcharts
http://www.iqchart.
com/iqchart/
From $35/month
including exchange fees.
IQ Chart Delayed Data
$24.95/month.
Excellent charts and the
competitive cost make
this a great chart service
for US stocks.
RealTick www.realtick.com. Real Tick Pro Plus
and the Realtick by
Townsend Anayltics
RealTick by Townsend
Analytics features
technical analysis,real-
time quote displays and
direct-access trading.
HotTrend provides real
time price alerts and
analysis.
StockPoint by
Streaming media
http://investor.stock
point.com
Free interactive OHLC bar
and candle charts with
indicators and moving
averages.
Excellent source of
equity charts covering
most exchanges
including the UK.
eSignal www.esignal.com eSignal provides the
data feed to interface
with numerous charting
packages,such as the
AGETreal-time package.
eSignal provides a real-
time data feed covering
European exchanges.
Technical
Package
Web Address Packages Comment
CycleTrader http://www.cycle-
trader.com/
CycleTimer $799.00. CycleTimer by Bradley F.
Cowan offers Time Price
vectors etc.
Extensive Cycle Research.
The site offers long term
data.
Gann Analyst. http://www.gann
alyst.com
Gann Analyst
Professional V3
A $ 695.00
Gannalyst Extended
$65.0
Gannalyst Lite Free
Free Data Downloader
and Data Converter.
Classic Gann type charts
and reasonable pricing
make this package
worthy of investigation.
Free quotes from Yahoo
for UK,US equities make
this a great introductory
package.
Web support.
Market Analyst http://www.market-
analyst.com
Gann analyst is the
additional module to
their Market Analyst
product expensive at
$1895.0
Market Analyst offers the
Gann
Analyst add in module.
Technical
Package
Web Address Packages Comment
Gannalyst http://www.gann
alyst.com
Gannalyst Lite.
Basic charting
A free basic package that
can be upgraded for $65
to obtain the free quotes
from Yahoo.
SpiffyCharts www.spiffycharts.
com
`SpiffyCharts 1.4.8
Basic chart package.
From the Behold
software group.
Supports CSI,MJK Dial
Data format data.
Technical
Package
Web Address Packages Comment
Financial Data
Calculator
www.futures-
software.com
FDC V1.2 $149.00 The Parabolic prediction
tool make this an
interesting program.
ELWave 6.2is an Elliott
Wave identification
program.RTanalysis Is
available with the Realtick
data feed.
Scanning Standard
Edition is $450.00
Qcharts by
Quote.com
www.quote.com
/quotecom/qcharts/
RT and Eod data service.
Excellent charting
interface.$79.95 per
month plus exchange
fees.
Great charts and user
interface make this a
popular real-time chart
and data site.
Now owned by Lycos.
MarketSmart by
Pcquote.com
www.Pcquote.com MarketSmart Charting
interface with a system
testing capability from
$49.95/month
Orbitis $75.00 per
month plus exchange
fees.
Hyperfeed Technologies
data transfer technology
for fast internet access.
Interfaces with a variety
of real-time packages
including Tradestation
and Excel.
Market Master http://www.easysoft
-inds.co.uk
MarketMaster 2000 has
great data handling and
all the standard tools.
MarketMaster was
originally released in
1986 and hence has
development history.
World market coverage
and equity sector
analysis make this an
easy to use package for
the analyst/fund
manager/equity
strategist.
Xpertrader www.XperTrader.
co.uk
Xpertrader. Net is a
complete charting TA
and decision support
system.
Built in trend
identification and
spectral analysis provide
some of the decision
support tools.
Issue 45 October 2002 MARKETTECHNICIAN 7
Free quote downloading software
Free Quotes Downloading Programs.Complete OHLCV EOD data can be
downloaded for the North American and European exchanges.Currently
Yahoo provides UK equity data as well.
The following downloaders interface with Yahoo and other free data
sources to collect free quotes:
Hquotes provides three interface programs at various levels.
From $55.00 http://www.hquotes.com/
Ashkon Downloader costs $39.95
http://www.ashkon.com/downloader.htmlDatashark.Downloader
http://datasharks.biz/Downloader_Info.php.
with a 10,700 quote library for $49.95
AnalyzerXl provides a free quotes via a Bulk Quotes Downloader.at
http://www.analyzerxl.com
Gannalyst Lite provides a good charting program and downloader
combined,all for $65 http://www.gannalyst.com
Software review for XLChartPro
Details:XL ChartPro $39 USD XL TradeLink $29 USD Bundle for $49 USD
Website: www.xlchartpro.iinet.net.au
Address:Penzina Pty Ltd,35 Prendwick Way,Willetton WA 6155 Australia
This is an intriguing piece of software which I have found useful in plotting
point and figure charts in Excel. The software is an add-in to Microsoft
Excel,so you need a PC system running this in order to get started. For
those of us who use point and figure charts,the last 15 years of advances
in computational power and charting systems often seemed to leave p&fs
out,or put them in as an afterthought. This has improved over the last few
years,but it can still be frustrating trying to plot p&fs in some of the major
systems. The XL ChartPro add-in fills in some of the gaps and,given the
reasonable cost and the data link facilities,it is worth considering.
The installation was fairly easy. I just downloaded the software (around 10
minutes on a slow line) and used the license key when prompted in order
to start up the program. The download came with two Microsoft Word
documents detailing how to use the XL ChartPro and Link add-ins. The
documentation was brief,easy to read and useful. The sample
spreadsheet was helpful in showing how to set up data sets and run the
XLChartPro program to plot the p&fs,with column counts appearing
below the chart and a volume histogram showing up below this. The
program has an automatic trendline plotting feature in which the
parameters (column count and penetration filter) can be adjusted by the
user. Moving averages can also be used,weighted by column count and
the software also plots volume weighted averages.
Was the software useful ? Yes,and the price makes it a bargain. If you like
p&fs this software fills a gap. Even our dealing room systems fail to plot
p&fs to our satisfaction,and I have been reduced to charting them by
hand now and then. The XL ChartPro plots our data (downloaded using a
DDE link from the dealing room systems) quite nicely and given a few
minutes of thought into data manipulation and naming conventions it is
quite easy to plot useful charts in Excel. The XL TradeLink add-in is a real
eye opener though,as this really exploits the power of the internet.
The Tradelink add-in can set up your Excel worksheet to link up to data
kept in the Ezy chart format (never heard of it) or Metastock (quite
popular) or text data files. Therefore,if you keep your data in those
formats,you can use it in Excel to plot p&fs. Even more impressive though
is the link to 50 exchanges. At the click of a button (around 5-10 seconds
on a slow line) the add-in downloads daily high,low,close and volume
data from the LSE or NYSE (or 48 other exchanges) for the specified share,
going back as many years as the exchange keeps it. For instance,I selected
the NYSE,typed in IBM,set the period for weeklyand the nearest date to
September 01,1982. The data downloaded in a few seconds. I then
opened up XL ChartPro and selected the IBM data sheet,and the chart was
plotted in seconds. While some exchanges keep their data in better
conditionthan others,for the majors this should not be a huge problem.
System limitations? None really. It mostly does what it says on the side of
the tin. I had to close down the XL ChartPro add-in and open it up a few
times when plotting some bond data downloaded from a dealing room
system,and I had to manipulate some FX data to get it to plot on a
reasonable scale,but these are easy problems to overcome (anybody using
Excel to model or manipulate data is probably used to this).
This software uses Excel to plot p&fs,giving the user the option to
manipulate the scales,the reversals,a percentage or normal scale and plot
moving averages or volume weighted averages on the chart while
displaying the column count and volume below the p&f chart. The data
link to the stock exchange websites is very useful,and this software
bundle should be attractive to any member who wants to look at their
share portfolio through the p&f charting technique without paying
exorbitant user fees for online systems,or who is dissatisfied with the way
that the more expensive systems plot p&fs. The only caveat is to make
sure that you have a robust PC (plenty of memory) as plotting charts is
addictive and you can easily use up a lot of memory on a large portfolio.
Gerry Celaya,Chief Strategist,Redtower Research
Joe Ross (1991) Trading is a business.
To quote the introduction Warning: this is a nasty book. It will take you
apart at the seams,point out your weaknesses as a futures trader and then
attempt to resurrect you as a self-disciplined person who can control his
trading in a business-like manner. If you cannot stand constructive criticism
or you do not wish to succeed as a futures trader,I suggest you lay this book
aside now and ignore it.
This book is aimed at people who have traded futures,albeit unsuccessfully.
The perspective is American and the examples given relate to the US
futures markets,however,the aspects discussed are universal and could be
applied in any trading environment.
The book is split into two sections; part one deals with the mental and
psychological factors that cause people to fail as traders and part two
details the mechanics of the correct trading methodology. Both money
management and trading techniques are described in detail.
Chapters four and six in part one are particularly enlightening because the
author challenges the reader to question their motives and expectations
for trading in general and the specific reasoning behind each trade.
Personality traits are identified by the common errors they induce and
solutions are suggested. The style of writing is thought provoking, if
nothing else. The methodology outlined in part two explains the
preparation work required before trading and exposes the myth that
futures trading is easy money with little effort.
Joe Ross comes from a family of traders and is quite critical of modern
technical indicators and the way people use them. Instead he focuses on
identifying congestion and trends, and on methods of trading these price
patterns.
Ross has a practical style of writing that stems from years of experience. His
exposure of traders mistakes makes essential reading for those with less than
a few years experience in a trading environment. I had to request this book
through my local library but there is a web site for anybody wanting to add it
to his or her personal library.http://www.ross-trading.com
This book is a refreshing change in that the author decries the use of
technical indicators/systems in favour of logical reasoning and good
money management. I recommend the book to anybody who is unfamiliar
with Joe Ross. The methods described are not a holy grail but an important
contribution to the subject of technical analysis.
Bruce Jamieson
The STALibrary has purchased a copy of this book.
Software review
By Gerry Celaya
Book review
MARKETTECHNICIAN Issue 45 October 2002 8
Candlesticks patterns generate effective short-term reversal signals,but
not all reversals are created equal. Because candlestick reversal patterns
range from one to three days,they can occur quite often. Not all reversals
turn into profitable trades and a second methodology can help filter less
robust signals.
The trend is your friend is a classic Wall Street adage and trading with
the trend can increase the chances of success. This may sound obvious,
but it is not always easy to participate once a trend is underway. Nobody
wants to be the last buyer in an uptrend or the last seller in a downtrend.
By combining trend identification with candlestick analysis,it is possible
to identify short-term entry points and trade in the direction of the larger
trend.
In this article,I will first identify some of the more robust candlestick
reversal patterns. Second,I will present a simple method to decide what
kind of reversals to look for and whento look. Third,I will provide some
examples that combine both techniques.
Narrowing the Candlestick patterns
In his bookCandlestick Charting Explained, Greg Morris tested over 50
candlestick reversal patterns based on prediction intervals of 3,5,7 and 9
days. The patterns that scored the best results across the board were:
Three Black Crows,Three White Soldiers,Three Inside Up,Three Outside
Up,Three Outside Down and Dark Cloud Cover.With the exception of Dark
Cloud Cover,all of these patterns consist of three candlesticks.
These results provide six patterns to work with. Three Black Crows and
Three White Soldiers are robust reversal patterns,but are relatively rare
and often skew the reward/risk ratio because of the considerable price
movement that has already occurred. In addition,each of the other
patterns has a two-candlestick equivalent that can provide an early alert.
For example:Three Outside Up is the same as a confirmed Bullish
Engulfing pattern and Three Inside Down is the same as a confirmed
Bearish Harami. Once the two-candlestick pattern emerges,traders can be
on alert for confirmation and act a bit quicker.As such,I would eliminate
Three Black Crows and Three White Soldiers.
I would then expand the remaining four patterns to include both the
bearish and bullish equivalents. Three Outside Up (bullish) corresponds
with Three Outside Down (bearish). Three Inside Down (bearish) would be
added to match with Three Inside Up (bullish) and the Piercing Pattern
(bullish) would be added to match with Dark Cloud Cover (bearish). The
Piercing Pattern and Dark Cloud Cover are two candlestick patterns and I
have added the third candlestick to show confirmation.
Identifying the trends
There are two trends that need to be established:the long-term trend and
the short-term. The long-term trend tells us whichcandlestick patterns to
consider and the short-term trend tells us whento look. Bullish
candlestick reversals are mandated when the long-term trend is up and
the short-term trend is down. Bearish candlestick reversals are mandated
when the long-term trend is down and the short-term trend is up. There
are many techniques for trend identification and I have chosen to use
moving averages for simplicity and illustrative purposes. Pattern
recognition,trendline breaks,resistance breakouts,support breaks,Elliott,
Gann and other systems are viable as well.
To determine the long-term trend and which candlestick reversals to
consider,I applied a 200-day simple moving average (SMA). The long-
term trend is considered bullish when a security is trading above its 200-
day day SMA and bearish when below. To determine the short-term trend
and when to look for candlestick reversals,I used a 10-day SMA.The short-
term trend is up when the prior days close is above the 10-day SMA and
down when below. It is worth emphasizing that bullish reversals can only
occur in short-term downtrends and bearish reversals can only occur in
short-term uptrends. Therefore,it is important to wait for the emergence
of a short-term trend that runs counter to the long-term trend.
AOL and a long-term downtrend
AOL first moved below its 200-day SMA in Feb-02 and then gyrated above
and below until the final break in Jul-01. The forays above the 200-day
SMA were brief and resulted in two whipsaws using these techniques.
Using Candlesticks to play the trend
By Arthur Hill
Issue 45 October 2002 MARKETTECHNICIAN 9
AOL was above its 200-day SMA and below its 10-day SMA when the
Three Outside Up patterns formed in May-01 and Jun-01. Both moves
were short-lived and AOL proceeded to break back below its 200-day SMA
in Jul-01. As entries were made soon after the short-term reversals,stops
would have been relatively close and losses minimal.
Once AOL moved below the 200-day SMA,it was time to look for bearish
reversal patterns and use the 10-day SMA for timing. The stock moved
above its 10-day SMA in Mar-02 and formed a Three Inside Down pattern
that foreshadowed a decline from 25.5 to 19. Notice that the Bearish
Harami was confirmed two days later instead of the very next day. A
second day confirmation is justifiable,but I would not extend the
confirmation period past the second day. Candlestick reversals are short-
term and should be confirmed within two days to be considered robust.
In May-02 and Jun-02,AOL remained below the 200-day SMA and made
two forays above the 10-day SMA. A Three Inside Down pattern formed in
mid May and again in mid June. Both marked the end of the latest
reaction rally and the resumption of the long-term downtrend to
foreshadow significant declines.
Intel and a long-term uptrend
INTC was one of the star Nasdaq performers from 1995 to 2000.With the
exception of a few brief dips,INTC remained above its 200-day SMA for
most time between Aug-98 and Aug-00. While below the 200-day SMA,
the stock did not produce any bearish reversal signals because of the quick
recovery. The trend was undeniably bullish from Aug-98 to Aug-00,but a
methodology was needed to decide when and how to jump on board.
In late 1999 and early 2000,INTC dipped below its 10-day SMA and forged
three candlestick reversals that provided entry opportunities. They were
all Three Outside Up patterns or Bullish Engulfings with confirmation. The
Dec-99 Three Outside Up formed as the stock tested the 200-day SMA.
Confirmation was on the second day,but came with a long white
candlestick to show underlying strength. The second Bullish Engulfing in
early January was confirmed with a gap up and doji. However,I would
consider this confirmation a bit suspect and difficult to act on. Doji signal
indecision and buying would have seemed like chasing the stock. The
third Bullish Engulfing was confirmed the very next day and
foreshadowed a move to the low 70s. For some addition confirmation,the
two-week pattern preceding the breakout looks like a falling flag and the
breakout signalled a resumption of the prior advance.
With INTC holding above its 200-day SMA in Aug-00,bullish reversals were
still the order of the day. After an extended move below the 10-day SMA,
INTC formed a Bullish Engulfing pattern and confirmed with an opening
gap and long white candlestick. The boost off support in the low 60s led
to a move above 75. Incidentally,the move above 75 marked the summer
high and the stock declined below its 200-day SMA in mid Sep-00.
Conclusion and summary
In a strong trend,it is often difficult to decide how and when to participate.
Once the long-term trend is established,looking for candlestick reversals in
the direction of that trend can offer an entry point with a good reward/risk
ratio. For long positions,exit points can be set just below the low of the first
candlestick in the reversal pattern. For short positions,the exit point can be
set just above the high of the first candlestick. Positions can be established
to catch 1-2 week moves or to partake in a longer ongoing trend.
Although I have used a 200-day simple moving average to establish the
long-term trend,other techniques can be used in conjunction with
candlestick reversals. After a strong advance,Fibonacci retracements (38%
or 62%) could be used to identify potential candlestick reversal points.As
broken support (neckline) often turns into resistance,a return to support-
turned-resistance can be used to anticipate candlestick reversals. The key
is to use candlestick reversals in harmony with the larger trend. Trading in
the direction of the long-term trend puts the wind at your back and can
enhance performance.
MARKETTECHNICIAN Issue 45 October 2002 10
My original article on Great Bear Markets was drafted in August 2001 in
response to a client question:How bad can it get,and when might it
finish? (see Market Technician Autumn 2001 or www.mtsresearch.com,
Articles section).Whilst the tragic events of September appeared to
increase the chances of an extended downtrend,the deep oversold
condition and the authoritiesresponse provided the stimulus for a strong
rebound. At the time,the action looked like a bear market rally,
particularly in the NYSE Composite (see Chart 1),but confirmation of that
only came in early May from the S&P 500 (Chart 2).
Against this background,I have been asked to write a follow-up article for
this issue. As a reminder,the table from the original article was as follows:
So how has market action conformed to the Great Bear Market (GBM)
model? In two ways it has not. Firstly,compare the percentage declines
in these four previous examples,when they were about five months off
their ultimate lows,with the World Index at 30th May.
Secondly,three out of the four previous GBMs all contained a Crash-type
decline in their first 10-15 weeks,whereas the current trend did not. But
note that even the UKs All-Share index,which began its decline quite
sedately in 1972,had caught up with the pack as the bear market neared
its final stages.
However,there are two similarities which have been sufficiently close that
it has been impossible to dismiss the argument that this is a GBM. Firstly,
if this is not a GBM,then it would be reasonable to expect that at some
point markets would produce a rally longer in time than the longest rallies
seen in the previous four GBMs. The longest two occurred in Gold and
Topix. Chart 3 below overlays the NYSE from September 2001 on the
longest GBM rallies in those two,all centred at the pre-rally lows. The
NYSE produced a classic GBM rally into its March 2002 high.
The second similarity is with Japans GBM,which had two important rallies
between its 1989 high and its 1992 low. The first occurred from the March
1990 low,retracing 50% of the initial decline (the US did the same
between November 1929 and April 1930). The second began at the end
of September 1990 and is the one illustrated in the above Gold and NYSE
comparison. It retraced 38.2% of the entire 1990 decline. The rallies in the
FTSE World Index from the Q2 and Q4 lows in 2001 also produced similar
retracements (Chart 4).
Further evidence of a Great Bear Market
By Peter Beuttell
NYSE COMPOSITE
S O N D J F M A M J J A S O N D J F M A M J
480
500
550
600
650
700
NYSE COMPOSITE - PRICE INDEX
Ratio
of
C to A
.618
1.0
Index intra-day
high
.618
1.0
b
a
A
c
B
a
b
Ratio of C to
A in Time
.618
.50
(B)
C
c
(A)
A classic Zigzag pattern, with an
Expanding Triangle B-wave. 1.0
Index Closing Closing % High Low Duration in
Peak Low Decline Date Date Days
1. S&P 500 * 31.83 4.41 86 6.9.29 8.7.32 1,035
2. UK All-Share 228.18 61.92 73 1.5.72 13.12.74 957
3. Gold 835 296 65 1.8.80 21.6.82 892
4. Topix 2,885 1,102 62 18.12.89 18.8.92 973
* : week ending figures and dates
S&P : February 1932 76%
FTASI : July 1974 56%
Gold : January 1982 55%
Topix : March 1992 51%
World : May 2002 29% )
These average
out at about 59%.
Very different in 2002.
}
Chart 1
S&P & 14-DAY RSI
M J J A S O N D J F M A M J
750
800
900
1000
1100
1200
1300
1350
10
20
40
60
80
100
120
140
160
180
190
S&P 500 COMPOSITE - PRICE INDEX
RSI MOMENTUM INDICATOR(R.H.SCALE)
Since 1969, whenever the S&P has been beneath its
flat or falling 200-day average, if a rally has failed
to produce a 70% 14-day RSI reading, and ...
70
30
Downtrend signal
... the RSI then hits
30% on the next
decline, that has
always been during an
on-going downtrend.
Chart 2
Gold: March
1980 - February
NYSE: September
2001 - June 2002
Topix: September
1990 - August 1991
Chart 3
JAPAN: 1990-92
1989 1990 1991 1992
1000
1200
1400
1600
1800
2000
2200
2400
2600
2800
3000
TOKYO SE (TOPIX) - PRICE INDEX
WORLD: 2000-02
1999 2000 2001 2002
170
180
200
220
240
260
280
300
320
340
360
FT WORLD(R.H.SCALE)
50% retracement
50%
retracement
38% 38%
Doing pretty much
what Japan did in
Fibonacci terms.
Similar rally
durations also.
Chart 4
Issue 45 October 2002 MARKETTECHNICIAN 11
What other evidence is there that a GBM-sized decline (62-86%) could yet
occur? Firstly,it has been clear since early 2001 that a large top pattern
was potentially in the process of forming.In the FTSE World Index,the
recent downward break targets anywhere from 140 down to the 90 area,
depending on whether you use the semi-log or linear measuring method.
That is a decline of 60-75% from the high (Chart 5).
There are always a number of Elliott Wave counts for any individual chart,
but the one illustrated in Chart 6 is preferred. One rule of using this
technique is that,once a 5-wave advance has completed,the ensuing
correction will retrace into the Area of the Previous Fourth Wave,and
often to its low point. The PFW is the 1990 decline and,if achieved,this
would represent between 57% and 70% off the 2000 high,which
compares with the 62-86% of previous GBMs.(MS World is used to
illustrate the count and PFW,due to its longer history).
Fibonacci targeting can also be used to arrive at a level which will help
indicate whether we should expect a fall greater than 50% from the high.
In an (A)-(B)-(C) bear market pattern,wave (C) is more normally related to
wave (A) in one of the following ratios:1.0/1.382/1.618. In the FTSE World
Index,this gives targets of 140/89/58. These represent 61/75/84%
declines from the high,and therefore confirm the risk of a GBM-sized
decline (compare these percentages with previous GBM actuals).
As I said in the original article,there is no way to prove that a GBM is in
progress until a lot of the damage is already done,but it was worth
publishing on a forewarned,forearmed basis. However,there is a further
problematic aspect to the current situation which was not obvious a
year ago.
The second table makes the point that the Time profile of this bear
market is very different from the four GBMs.They are about intensity
a large percentage fall in a given timeframe. There are certainly others
which have lasted longer,and the US bear market of 1937-1942 is one
example. However,the problem is that if this is a trueGBM (i.e.it sticks to
the 900-1,000 day timetable) the inescapable conclusion is that a final
decline,which takes markets 62-86% off their 2000 highs,will be
sandwiched into the four-month period centred on 13th November 2002,
with a point target of 6th November when the current Cycle of Pi bottoms
(see the first article for the time aspect). Whereas three of the previous
GBMs began with a Crash-type or sustained decline,on this occasion the
profile will reverse,and this GBM will end with one. Chart 7 below
illustrates that Time and Price target zone.
This article was drafted in June/July,and I am adding this paragraph to the
final proof in very early September. At this stage,I am pretty convinced
that we are eventually going to see a decline of GBM degree. Since the
downside wave pattern which began in March 2002 is incomplete,and
the July-August recovery looks like another completed bear market rally,
the final decline could even have begun. Will the bear market become a
trueGBM? I cannot say,but I fear it will. I expect to see the S&P 500 trade
beneath 600,and possibly as low as 430,at some point in the next two to
fifteen months,based on Fibonacci time and price projections. Chart 8
illustrates an interesting coincidence in this respect.
If the worst happens,at least there are these previous GBM and other
examples to look at to tell us what indicator readings to expect at the low,
and what market action is likely to be in the immediate aftermath. Let us
hope it does not come to that.
Peter Beuttell is a Director of MTS Research Ltd.,which provides market and
stock timing advice to institutional investors. See www.mtsresearch.com and
www.taranalysis.com
FTSE WORLD INDEX (LOCAL)
87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02
90
100
150
200
250
300
350
400
FTSE W WORLD - PRICE INDEX
The semi-log scale measurement
projects to the 140 area, a 60%
decline from the high. This pattern
requires volume confirmation, which
has been given in the NYSE, and is
acceptable in view of the 50%+
weighting of the US in this index.
S
H
S
Chart 5
FTSE WORLD INDEX (LOCAL)
90 91 92 93 94 95 96 97 98 99 00 01 02 03 04
50
100
150
200
250
300
350
400
FTSE W WORLD - PRICE INDEX
Note how the % retracements off
the high tie in fairly closely with
the ratio targets for wave (C).
The equivalent GBM lows are
listed on the left.
(A)
(B)
(C)
Tokyo 1992
London 1974
New York 1932
.618 @ 136
1.0 @
140
.7236 @ 84
.854 @ 52
1.382 @ 89
1.618 @ 58
GBM time zone 02.09.02 - 21.01.03
Ratios of
(C) to (A)
Chart 7
(I)
(II)
(III)
(IV)
(V)
It may be a coincidence, but
around 6th November, the
trend channel lies in the area
of the .7236 retracement off
the high at 429.
S&P 500 Quarterly since 1927
1.0
1.0
(V) = (I) + (III)
Chart 8
MS WORLD INDEX (LOCAL)
73 74 75 76 77 78 79 80 81 82 83 84 85 86 87
60
200
400
600
800
1000
1200
MSCI WORLD - PRICE INDEX
88 89 90 91 92 93 94 95 96 97 98 99 00 01 02
60
200
400
600
800
1000
1200
1
2
3
4
5
Previous Fourth
Wave support and
target area.
Chart 6
MARKETTECHNICIAN Issue 45 October 2002 12
It is 20 years since the personal computer was first used to bring Technical
Analysis to the masses. In fact,in those early days,it was felt that Technical
Analysis was far too sophisticated for a computer to handle and many
traditional Technical Analysts refused to use computers and perhaps rightly
so. The graphics capabilities of the IBM PC rendered computer-drawn
charts almost unreadable,but it was a step in the right direction. The
speed with which those early programs drew charts would be regarded as
laughable today,but early Technical Analysis programs which took 12
seconds to draw a bar chart and a moving average had to be compared
with the hour or more it took to generate the same chart by hand. Being
able to change a moving average or oscillator period and see the result
allowed more research to be undertaken. It opened up a whole world of
calculated indicators which would otherwise have been impossible.
For many years after those first programs appeared,most professional
Technical Analysts continued to use hand-drawn charts as well,but software
companies persevered and hardware manufacturers soon realised that
computers were being used for more than just word processing.
Consequently graphics resolutions improved and are a state now where
hand-drawn charts can no longer compete. Thats the good news.
The bad news is that the post-computer age allowed anyone with a
computer to be a technical analyst (lower case intentional). That in itself is
not a bad thing. We welcome the wider use of our discipline,but what is
worrying is that in most cases this computer-generation of technical
analysts have no understanding of the charts,no knowledge of what
technical analysis is all about and have blind faith in lines the software is
producing for them. Simply clicking on a button allows the most
sophisticated calculation to be performed in an instant and the resultant
chart to be drawn.
The problem with this is the assumption that the chart is technically correct
calculated correctly and plotted correctly. Without any background
knowledge of the chart or its history or experience of manual plotting,the
user has no way of knowing if the chart is right. Those of us who used to
draw charts by hand,know that there is more to technical analysis than just
drawing a chart. We know that we have to become intimately involved
with the chart and understand what it is trying to tell us. In becoming
intimate,we would instantly recognise an incorrect plot or miscalculation.
It is unlikely however that any of us would ever wish to go back to those
pre-computer days. Now,we can do so much more than we used to.
Where constructing charts took 90% of our time,construction now takes
no time at all,allowing us to spend all our time on analysis instead.
Many traditional charts,however,have suffered at the hands of software
developers,to such an extent that users of them now think they are plotted
correctly,when in fact they are not. Continued use of the wrong method
does not make it right. Welles Wilders popular RSI indicator has suffered
because somewhere along the line,a computer programmer (not a
Technical Analyst),without access to Wilders original work,misunderstood
his idiosyncratic but effective averaging technique and assumed a simple
arithmetic or exponential averaging technique instead. The result is a chart
that looks like an RSI but isnt one. Its no wonder that two analysts
comparing their RSIs cant understand why they get different results. The
same scenario applies to many of the well-known calculated indicators.
But one chart that has suffered more than most from mis-plotting is the
Point & Figure chart and in particular the original 1-box reversal chart.
Modern books tend to cover 3-box reversal charts only. It is rare to see
1-box mentioned and when it is,it is often shown incorrectly.
If the user of software does not understand how these charts were
developed and hence constructed,it is impossible to know whether the
Point & Figure chart on the screen is right or wrong,with damaging
implications.
Most users of Point & Figure know in the back of their minds that every
time there is the appropriate reversal,a column of Xs changes to a column
of Os and vice versa. You will see authors state that there can never be an O
and X in the same column. But is this always the case? With 3-box reversal
charts it is. If a rising column of Xs reverses by the value of 3 boxes (it cant
reverse by less),you must move diagonally across and plot a column of Os
in the opposite direction (figure 1). If it then reverses back again by the
value of 3 boxes,you move across diagonally and plot a column of Xs
(figure 2). This is the Point & Figure Chart that most are familiar with.
But why do we move across to a new column when
we change from Xs to Os or Os to Xs? The reason is
simply that the only empty squares are in the next
column and so one has to move across to fit in the
next column of Xs or Os. So,the books are correct,
you may never have an O and X in the same column.
This is however not the strict rule that so many
believe. The 3-box reversal chart is a compressed
version of the original 1-box reversal chart. With the
1-box chart it is not always necessary to move across
to the next column,because in certain circumstances
there is an empty box in the current column.
Consider this scenario. If a rising column of Xs
reverses by the value of 1 box,you must move
diagonally across to the next column to plot the O,
because the box in the current column is already
occupied. (figure 3).
But what if the next price reverses back up again by
the value of at least 1 box? You do not have to move
across one column to plot the X,because there is an
empty square into which the X may be placed.
(figure 4) This means that you canhave an X and O
in the same column. Figure 5 shows the correct way
to account for a reversal of 1-box and a subsequent
reversal back again. Figure 6 shows the incorrect
method of plotting the same chart. Notice how a
new column is started each time there is a reversal.
Other than ignorance,there is no dispute as to which
is the correct method. Reference to original works
will confirm it immediately,so why do we see the
incorrect method so often? The answer is simple.
Most users of Point &
Figure have had the no X
and O in the same
columnrule drummed
into them. Consequently
software developers and
even some authors have
followed the same rule
and suddenly the correct
plotting method
becomes clouded and
may even be lost.
Does continued use of the wrong method
make it right?
By Jeremy du Plessis
Figure 1
Figure 2
Figure 3
Figure 4
x
x
x
x
x
o
x
x
x
x
x
x
o
x
x
x
x
x
x
o
x
o
x
o o
o
x
o
x
o
x
o
x
x
x
x
x
o
Figure 5
Figure 6
x
x
x
x
x
o
x
o
x
o
x
o
o
x
o
x
o
x
o
x
x
x
o
x
x
x
x
x
x
x
o
o
o
x
x
x
x
x
o
o
o
x
x
x
Issue 45 October 2002 MARKETTECHNICIAN 13
To fully understand the logic of Point & Figure
charts and how they should be plotted we need to
go back to the late 19th century. Traders wanted a
way of recording price changes quickly and easily
whilst standing on the trading floor or in the
reception area of the brokers office. So,they
started writing down prices in columns as the stock
traded up or down. Each time a new price was
recorded,it was written into a square. The only
reason for moving across one column was to avoid writing a figure on top
of another. If the price recorded was a series of prices,such as 20,21,22,
23,24,23,24,25,24,25,the first five figures would be written down in a
column rising as 20,21,22,23,24. Then the price records 23. Because the
square below 24 already has the figure 23 in it,the trader moves across to
the next column and writes in 23.The price then records 24. The square
above 23 is unoccupied,so 24 may be written in the square above without
having to move across. The next price is 25. The square above is
unoccupied so 25 is written above the 24. The price then falls to 24. The
24 square in the same column is already occupied so the trader must
move across again to plot 24. The price rises to 25. The 25 square in the
same column is unoccupied,so 25 may be written in the square above 24.
And so the Point & Figure Chart was born. (figure 7). The straight line
overlay shows the sequence in which the numbers are written down.
As time passed,traders became tired of writing down
numbers and so they plotted Xs instead,placing
numbers at the 5 and 0 levels. (figure 8) Having Xs in
both up columns and down columns was confusing
and so Xs were used for up columns and Os for
down columns.(figure 9) Referring to figure 9,it is
easy to see now that it is possible,and makes sense,
to have an O and X in the same column.
When A.W.Cohen wrote his excellent work on
3-box reversal Point & Figure charts in 1968, it did
not cover the original 1-box charts. Many analysts
switched to 3-box charts and so began the demise
of the 1-box reversal. When Technical Analysis
software started appearing,1-box charts were
mostly ignored and when they did appear they
were often plotted incorrectly because the correct
method had been forgotten through lack of use.
These incorrect charts were then inadvertently
used by those new to Technical Analysis and when books were written,
everyone began to think the wrong method was right.
The question is does it matter? Yes,it does,for a number of reasons. This
charting method was created over 100 years ago. Considerable time,
effort and experience has gone into studying these charts over this time.
It is important that we carry forward these original works and methods
otherwise we can claim no history to our discipline.
Congestion plays an important role in Point & Figure analysis. We can read
so much into the shape of the patterns created by the Xs and Os. The
width of congestion patterns are used for establishing price targets. But if
the congestion includes a number of 1-box price changes,the incorrect
method will yield a much wider congestion area and hence an excessive
price target because the target is established by counting the number of
columns in the congestion area. The incorrect method will yield more
columns because a new column is started each time there is a reversal.
Trends and trend lines are important in all charts and none more so than
Point & Figure charts. A reversal of 1-box followed by the immediate reversal
back again (figure 4) shows strength in a trend. Because the O and
X are in the same column,the angle of the trend line is steeper. Consequently,
the position and angle of trend lines will be different if the incorrect method
is used. This is more so when tick data is used to plot the Point & Figure chart,
remembering,however,that genuine Point & Figure Charts require tick data.
With 1-box charts,repeated recording of the prices in a narrow range
1-box size apart shows resistance or support building up. These clusters
of Os and Xs sittingon top of one another are like building bricks (figure
5). Its almost as if the O below the X is a supporting brickunderpinning
the X. In many cases you will see a staircase being created as the price
forms a new column,then steps backby one O and then resumes the
trend when an X is placed on top of the O. The converse is true in down
trends,where the lone X,provides a lidon top of the O. This visual aspect
is lost completely if the incorrect method is used.
It is easy to see the differences and readability by comparing the two
charts below. Chart 1 shows a 1x1 Point & Figure chart of the FTSE Future
constructed with tick data using the correct plotting method. Chart 2
shows the same chart using the incorrect method.
Chart 1
Compare the up trend in both charts. Notice how,in the correct chart,the
Os provide the underpinning during the up trend as if the price is
climbing a staircase. This is not apparent in the incorrect Chart 2.
Chart 2
Notice also the narrow 1-box trading ranges created during the up trend
and how they cluster together showing short-term support and
resistance. Chart 1 shows these far more clearly than Chart 2.
During the up trend the price traded in a number of narrow trading
ranges,building up strength for the assault to the next congestion area,
almost like the next landingon the staircase. By measuring the width of
these congestion patterns,1-box charts allow us to estimate where the
next congestion level will occur. The counts are all taken at the base of
the break out column. In each case the incorrect method overstates the
count. This overestimation is more exaggerated,the more 1-box reversals
there are within the congestion pattern. It becomes even more apparent
when a congestion area has taken many weeks or months rather than a
few minutes to form and is then used to count the next major level.
In spite of the evidence,there are those who will argue that the incorrect
1-box method is better. Many will argue that the RSI calculated with an
exponential average gives better results than one calculated using Wilders
averaging technique. There should be nothing to prevent Technical
Analysts from using the incorrect method or modifying any chart,provided
that it is done with full knowledge and understanding and that the user of
the incorrect method states that it is a non-traditional or modified method.
All Technical Analysts should know the right and the wrong method. Ours is
not a young discipline. We have a vast body of written knowledge to fall
back on when in doubt. It is important that,as professionals,we are
perceived to know what we are talking about and that there is a formal and
correct way of drawing all the charts that we use. There are those opposed
to our discipline that are waiting to catch us out. We must educate and
preserve the charting methods upon which Technical Analysis has been built
in the last 100 years. Continued use of the wrong does not make it right.
Jeremy du Plessis CMT,is head of Technical Analysis at Updata plc and
developed one of the first PC based Technical Analysis software systems 20
years ago. He teaches the Point & Figure module for the STA diploma.
Figure 7
Figure 8
5
x
x
5
x
24
23
22
21
20
25
24
23
25
24
Figure 9
x
x
x
x
x
x
x
o
x
o
x
x
x
x
0
MARKETTECHNICIAN Issue 45 October 2002 14
Introduction
One of the most difficult questions facing a technical analyst is how to
differentiate a signal that requires a minor portfolio adjustment from a
signal that necessitates a major re-appraisal of a strategic view. The
traditional answer has been that it doesnt matter insofar as accurate
short-term decisions will ensure that the long-term decisions are largely
unnecessary. This answer is probably correct for day-traders and small-
scale investors. Equally,however,it is largely incorrect for strategic
investors and large-scale money managers:the movement of large sums
of money can incur penal transactions costs,is often disallowed under
management agreements,and inevitably exposes funds to huge
performance risks against competitors. Consequently,even the most
enthusiastic of technical fund managers will cast more than a passing
glance at so-called economic fundamentals.
This and the fact that technical analysis still does not have a satisfactory
theoretical underpinning amongst academics helps to explain why
economic analysis continues to have such an intractable hold over
investment decisions. Of course,it wouldnt matter if economic forecasting
was accurate but it isnt. First,the intellectual framework of economics is
still largely incapable of placing accurate time frames on turning points in
relevant variables. Second,economic forecasting is based on data that is
both out-of-date and inaccurate. Indeed,(and I suppose Im allowed to say
this in a market techniciansjournal) at the point where major investment
decisions are likely to be made,economic analysis often provides no more
than an intellectual comfort for an emotionally based guess.
So were left with a situation where technical analysis is often not clear about
the importance of its signals and where economic forecasting is often not
clear about the imminence of a turning point. Is there any way that the
resulting uncertainty can be reduced? In my opinion,there is. It still needs a
great deal of research and thought but,in principle,the solution (or,perhaps,
asolution) lies in applying some of the concepts of technical analysis to
economic variables. The essential point is that financial market fluctuations
and economic oscillations are both products of group behaviour. As such,
both will reveal non-random patterns and rhythmic vibrations. Moreover,the
two arenas will necessarily be integrated in some way. Specifically,economic
behaviour provides the context within which financial markets move,and
price trends in equities and bonds will generate feedback into the economy.
All that is required is that we know what to look for.
The theoretical background
I have for some years been working with the hypothesis that the conflict
between random fluctuations and ordered process in markets is resolved
by the influence of the group. That is,the inherent desire of otherwise
independent individuals to participate in greater wholes,allows the
wholeto devolve order onto the participants. The influence of the whole
is psychological,emotional and,accordingly,powerful:individuals are
eventually induced to do what everyone else is doing. This is why large
numbers of people simultaneously buy dotcom stocks or second houses
to rent,when reason suggests that caution might be more appropriate.
So,if we can allow that each person has a tendency both to be an individual
and to be part of a group,then two of the fundamental assumptions of
economic theory fall flat on their face. First,aggregate behaviour is not just
the sum of the parts;it is something significantly more. Second,this
augmented behaviour involves an important non-rational (and sometimes
even irrational) dimension. In effect,what economic theory ignores is that
each of us has an impulse to create meaningfor ourselves,where meaning
is measured by our inner world of feelings,and where feelings are
stimulated by,and integrated with,our relationships with others.
All this is now pretty standard psychology,but many may still ask:so
what? Why does it matter to forecasting that people have feelings and
have a tendency to do things together? Specifically,doesnt the
simplifying assumption of rational behaviour by independent individuals
just make the forecasting models easier to use? Actually,the literal
answer is yes. But,unfortunately,it also makes the forecasts less accurate.
Critically,the simplifying assumption of economics does not allow for the
simple reality of human behaviour that lies at the heart of cyclical
behaviour the emergence of satiation. Indeed,in economics,satiation is
treated as an aberration:the unspoken presumption is that we never have
enough goods(note the terminology). Hence,growth can and should
continue persistently and indefinitely,subject only to the birth rate (which
creates new entrants to the consumer market) and the ability of the
system to generate new goods(through innovation).
In nature,however,every living system oscillates between a state of
activity and a state of rest. Indeed,rest is essential in order to re-energise
an organism and delay entropy. So,satiation with activity causes an
organism to switch from activity into rest and,conversely,satiation with
rest causes an organism to switch from rest into activity. Hence,for
example,as we approach satiation at the end of a day,we begin to lose
energy,our body heat drops and our metabolism slows. Then,suddenly,
we fall asleep. We traverse a process gap. Conversely,when weve had
enough sleep,our energy potential is re-established,so our organ systems
start to speed up and our body temperature rises. Again,we experience a
process gap and,suddenly,we are awake. Every living organism regularly,
and at some infinitesimal moment in time,passes over the cusp that
delineates sleep from rest. Moreover,these moments in time are entrained
across nature. Most of us,for example,go to sleep before midnight and
wake up in the early morning (give or take a late night out!). The
harmonisation of activity and rest throughout the biosphere reduces
interference and allows total energy usage to be minimised.
There is another element here, which is also critical to our understanding
of oscillations in living systems. Living systems respond to the input of
energy and information. The interesting point is that when a system
receives new energy or information, it has to divert existing energy to
deal with it. A simple example is eating. Food provides new energy to
our bodies. The first thing that happens, however, is that energy
resources are diverted to deal with the transformation and absorption of
the food. That is, our freely available energy tends to drop and, while this
is happening, we actually tend to slow down and feel sleepy. In a sense,
this slowdown revisits the drop in energy that initially signalled the need
to eat. But there is a huge qualitative difference: the slowdown without
food marks the end of a phase, while the slowdown with food marks the
transition to a new phase. Having eaten, our energy is eventually
considerably enhanced and we can go about our activities with some
degree of vigour.
Obviously what we have here is a three-wave pattern that proceeds
through input-absorption-application. Importantly,this pattern does not
just apply to energy;it also applies to information. In human beings,it has
specifically been found
1
that the diversion of energy to deal with the
transfer of information from short-term memory to long-term memory
momentarily interrupts the ability to apply new learnings. So,the three-
wave pattern is also the signature of learning. The first two stages are the
true learning stages (where information actually alters the qualitative
structure of the organism) and the third stage constitutes the
confirmation that learning has occurred.
There are thus four ideas that can be used to produce a model of oscillations
that mirrors what happens in financial markets. These ideas are:(1)
individuals combine into groups,(2) groups respond to the input of energy
and information from their environment,(3) the associated absorption of
new energy and information involves a pause in activity,and (4) satiation
causes a polarity switch between activity and rest. One inference is that a
financial market group can be treated as if it were a living organism.
Financial market oscillations
At the final low of a financial market cycle,the bears will be overextended
and the market will be oversold. In the background,we can hypothesise
either that fundamentals may have started to improve or that the market
has over-discounted the fundamentals. The market will be actively falling
but either there will be no new bearish information to sustain the drop,or
the number of investors willing to sell will dry up. So the bear will run out
of energy (in terms of information and financial allocations) and the
situation will be ripe for a bear squeeze. When this bear squeeze occurs,
Pattern and periodicity in financial cycles
By Tony Plummer
Issue 45 October 2002 MARKETTECHNICIAN 15
it will actually develop some momentum and go further than the
fundamentalists either expected or wanted. Note that the squeeze has
arisen out of satiation ie,the bear case has been taken as far as it can go.
It therefore reverses the polarity of the market from bearish to bullish. It is
the financial equivalent of waking up in the morning.
On this analysis,the bear squeeze is a piece of information that the market
has to digest. There will be a resistance level beyond which investors will
be unwilling to take the market because they are unsure. Most will not be
able specifically to identify this level (although it may be quite discernible
in mathematical terms
2
). However,there will be a sense that the market
has risen too far. Nevertheless,much time and effort will now be taken in
re-analysing fundamentals,or waiting for fundamentals to catch up.
During this phase,therefore,energy is diverted to absorbing the
implications of the bear squeeze and the market drops to re-test the low.
Many will,of course,see the re-test as being a renewal of the bear. However,
not only has the energy available for the bear weakened,but bullish energy
has been bolstered. So,as the market re-tests the low,volume,momentum
and open interest may contract. There may even be mathematical limits to
the extent of the drop. At this stage,the market is absorbing the information
ie,learning that the market may be reversing trend.
Eventually,trading selling dries up and the market stabilises. It is at this
point that the market is peculiarly vulnerable to bullish information from
the environment. It may take only one generally available item of such
news,or a small amount of persistent buying by those who believe that
they can anticipate the news,to generate a sharp upward impulse move.
The market has now learnt that it is bullish and most subsequent
incoming data will merely confirm that this is so. Negative news creates
no more than counter-trend setbacks and buying opportunities.
Ultimately,the market becomes satiated,in the sense that investors have
enough stock and a shortage of buyers develops. The market is therefore
overbought and vulnerable to profit taking. If fundamentals have started
to turn negative,or if the market has over-discounted the fundamentals,the
profit taking could act as the information shock that reverses the polarity of
the market from bullish to bearish. Then the reverse process sets in.
The price pulse
This model is,of course,very simplistic. Nevertheless,it gives us the basis
of a cycle that not only has an understandable mechanism but also has a
very specific pattern. In fact,one complete beat of the cycle has six
phases three waves up and three waves down. The first two waves of
each movement are a true learning phase,consisting of an information
shock and its absorption. The third wave is the genuine impulse wave
that applies the learnt information. In Figure 1 below,the upwaves are
denoted 1-2-3 and the downwaves are denoted A-B-C.
Figure 1:The cycle mechanism
It is my observation that this pattern underlies all market movements.
3
Moreover,it provides an answer to those who have either found market
cycles to be too arrhythmic or believe regular market fluctuations are
inconsistent with the assumption of creative behaviour by market
participants. First,the cycle is defined by the patternof internal change,
not by the precision of the periodicity. Second,the pattern is an essential
part of the process of creative adjustment to external change. All that we
need to do is isolate the three-wave patterns inherent in the up-phase
and down-phase of a cycle.
The 11-year cycle in the Dow
In order to provide a basis for analysis,it might be useful to look
straightaway at some practical examples. Space limitations mean that we
can only look at a tiny part of the massive amount of data available. I shall
concentrate on the US Dow Jones Industrial Average. The evidence
which is borne out by the detailed work of many others indicates that
one of the dominant cycles in this average has a duration of (roughly)
11 years. So major price lows since the end of WWII are taken to have
occurred in autumn 1946,winter 1957,summer 1970,autumn1981,
autumn 1990 and summer 2002.
4
Figures 2-i to 2-v below show the 6-month rates of change in the monthly
closes of the Dow. Each chart spans a period of approximately eleven
years,measured from momentum trough to momentum trough. Figure
2-i covers the 133 months from November 1946 to December 1957,Figure
2-ii covers the 150 months from December 1957 to June 1970. Figure 2-iii
covers the 135 months from June 1970 to September 1981. Figure 2-iv
covers the 110 months from September 1981 to November 1990. And
Figure 2-v covers the 140 months from November 1990 to July 2002.
Figure 2-i:DJIA,November 1946 to December 1957
Figure 2-ii:DJIA,December 1957 to June 1970
Figure 2-iii:DJIA,June 1970 to Sep 1981






































-35
-25
-15
-5
5
15
25
35
N
o
v
-
4
6
M
a
y
-
4
7
N
o
v
-
4
7
M
a
y
-
4
8
N
o
v
-
4
8
M
a
y
-
4
9
N
o
v
-
4
9
M
a
y
-
5
0
N
o
v
-
5
0
M
a
y
-
5
1
N
o
v
-
5
1
M
a
y
-
5
2
N
o
v
-
5
2
M
a
y
-
5
3
N
o
v
-
5
3
M
a
y
-
5
4
N
o
v
-
5
4
M
a
y
-
5
5
N
o
v
-
5
5
M
a
y
-
5
6
N
o
v
-
5
6
M
a
y
-
5
7
N
o
v
-
5
7
6
-
m
o
n
t
h

%

c
h
a
n
g
e
-35
-25
-15
-5
5
15
25
35
0 6 12 18 24 30 36 42 48 54 60 66 72 78 84 90 96 102 108 114 120 126 132
Months since low
6
-
m
o
n
t
h

%

c
h
a
n
g
e
1
C
B
A
3
0 2




















-35
-25
-15
-5
5
15
25
35
D
e
c
-
5
7
J
u
n
-
5
8
D
e
c
-
5
8
J
u
n
-
5
9
D
e
c
-
5
9
J
u
n
-
6
0
D
e
c
-
6
0
J
u
n
-
6
1
D
e
c
-
6
1
J
u
n
-
6
2
D
e
c
-
6
2
J
u
n
-
6
3
D
e
c
-
6
3
J
u
n
-
6
4
D
e
c
-
6
4
J
u
n
-
6
5
D
e
c
-
6
5
J
u
n
-
6
6
D
e
c
-
6
6
J
u
n
-
6
7
D
e
c
-
6
7
J
u
n
-
6
8
D
e
c
-
6
8
J
u
n
-
6
9
D
e
c
-
6
9
J
u
n
-
7
0
6
-
m
o
n
t
h

%

c
h
a
n
g
e
-35
-25
-15
-5
5
15
25
35
0 6 12 18 24 30 36 42 48 54 60 66 72 78 84 90 96 102 108 114 120 126 132 138 144 150
Months since low
6
-
m
o
n
t
h

%

c
h
a
n
g
e
1
C
B
A
3
2 0




















-35
-25
-15
-5
5
15
25
35
45
J
u
n
-
7
0
D
e
c
-
7
0
J
u
n
-
7
1
D
e
c
-
7
1
J
u
n
-
7
2
D
e
c
-
7
2
J
u
n
-
7
3
D
e
c
-
7
3
J
u
n
-
7
4
D
e
c
-
7
4
J
u
n
-
7
5
D
e
c
-
7
5
J
u
n
-
7
6
D
e
c
-
7
6
J
u
n
-
7
7
D
e
c
-
7
7
J
u
n
-
7
8
D
e
c
-
7
8
J
u
n
-
7
9
D
e
c
-
7
9
J
u
n
-
8
0
D
e
c
-
8
0
J
u
n
-
8
1
6
-
m
o
n
t
h

%

c
h
a
n
g
e
-35
-25
-15
-5
5
15
25
35
45
0 6 12 18 24 30 36 42 48 54 60 66 72 78 84 90 96 102 108 114 120 126 132
Months since low
6
-
m
o
n
t
h

%

c
h
a
n
g
e
1
C
B
A
3
2
0
MARKETTECHNICIAN Issue 45 October 2002 16
Figure 2-iv:DJIA,September 1981 to November 1990
Figure 2-v:DJIA,November 1990 to July 2002
What is apparent from these graphs is that,although there are distortions
and anomalies that need to be explained,the 11-year cycle does indeed
appear to contain a basic 6-wave (1-2-3 up/A-B-C down) pattern.
Triadic cycles
If the pattern is so important,it is worth considering what characteristics a
complete cycle might have. The starting point is to see if there are any
fixed relationships between successive troughs and successive peaks.
Table 1 below measures the time elapse,measured in months,between
important turning points within each of the 11-year cycles. The lows are
absolute momentum lows. Highs are momentum highs,but some of
them are relative rather than absolute. That is,some of the highs relate to
price highs that are generated on lower momentum (a non-
confirmation). This last point will be covered in a little more detail below
under the section entitled Biases in momentum.
Table 1:Internal cycle timings
One of the immediate observations is that,although there is plenty of
variability,the positioning of the main peaks and troughs conforms to a
very simple formula. This is that the major troughs occur at around 33%
and 66% of the total time elapse of the average cycle and that peaks
occur at around 17%,50% and 83% of the average cycle.In other words,
the cycles are divided into balanced triads:the main cycle consists of
three lower-level cycles,each of which has a duration that is about a third
of the duration of the main one. This is consistent with the operation of a
cycle of about 44 months,which in turn correlates with the short-term
business cycle found by Joseph Kitchin. Further,each of these short-term
cycles tends to register an important peak very close to its own mid-point.
There is variability here,but not randomness.
Figure 3 below shows an idealised pattern for the cycles. The higher-level
cycle is represented by the move from 0 to C ie,1-2-3 up and A-B-C
down. It consists of three lower-level cycles,each of which itself consists
of three cycles. Significant lows occur one-third and two-thirds along the
time elapse of the higher-level cycle. Important highs occur at one-sixth,
one-half and five-sixths along the time elapse of the cycle. And the mid-
point of the second lower-level cycle will coincide with the mid-point of
the overarching higher-level cycle.
Figure 3:The cycle model
Cycle translation
There are,of course,variations,which are based on the essential creativity
of markets. There are two major types of distortion that can arise. The
first is that the peaks in lower-level cycles may be biased through time
either by their relative positionwithin the larger degree cycle or just by the
power of the larger degree cycle. These biases are often called
translations. In principle,each cycle within a triad is likely to have certain
characteristics. The first cycle is the basecycle,which starts the big cycle
off. It can often be centred rather than biased. The second cycle is likely
to be the trendcycle,which provides the main impulse move. As such,it is
likely to be rightward biased such that it peaks late. The third cycle in a
triad is,accordingly,the terminal cycle,which unwinds all the excesses of
the trend cycle and incorporates the main bear phase. It would tend to be
leftward biased such that it peaks early. Readers may recognise that this
cartography reflects R.N. Elliotts insights.
These tendencies are modelled in Figure 4 below. Here,the dashed line
indicates the existence of the higher-level cycle (although not its locus),
which is divided into three lower-level cycles. The latter are themselves
sub-divided into triads. The translations are measured either in relation to
the 17%,50% and 83% time divisions of the relevant higher-level cycle or
in relation to the 50% time division of the lower-level cycle itself.
Figure 4:Translations within the triad





















-35
-25
-15
-5
5
15
25
35
45
S
e
p
-
8
1
J
a
n
-
8
2
M
a
y
-
8
2
S
e
p
-
8
2
J
a
n
-
8
3
M
a
y
-
8
3
S
e
p
-
8
3
J
a
n
-
8
4
M
a
y
-
8
4
S
e
p
-
8
4
J
a
n
-
8
5
M
a
y
-
8
5
S
e
p
-
8
5
J
a
n
-
8
6
M
a
y
-
8
6
S
e
p
-
8
6
J
a
n
-
8
7
M
a
y
-
8
7
S
e
p
-
8
7
J
a
n
-
8
8
M
a
y
-
8
8
S
e
p
-
8
8
J
a
n
-
8
9
M
a
y
-
8
9
S
e
p
-
8
9
J
a
n
-
9
0
M
a
y
-
9
0
S
e
p
-
9
0
6
-
m
o
n
t
h

%

c
h
a
n
g
e
-35
-25
-15
-5
5
15
25
35
45
0 6 12 18 24 30 36 42 48 54 60 66 72 78 84 90 96 102 108
Months since low
6
-
m
o
n
t
h

%

c
h
a
n
g
e
1
C
B
A
3
2 0



















-20
-10
0
10
20
30
N
o
v
-
9
0
M
a
y
-
9
1
N
o
v
-
9
1
M
a
y
-
9
2
N
o
v
-
9
2
M
a
y
-
9
3
N
o
v
-
9
3
M
a
y
-
9
4
N
o
v
-
9
4
M
a
y
-
9
5
N
o
v
-
9
5
M
a
y
-
9
6
N
o
v
-
9
6
M
a
y
-
9
7
N
o
v
-
9
7
M
a
y
-
9
8
N
o
v
-
9
8
M
a
y
-
9
9
N
o
v
-
9
9
M
a
y
-
0
0
N
o
v
-
0
0
M
a
y
-
0
1
N
o
v
-
0
1
M
a
y
-
0
2
6
-
m
o
n
t
h

%

c
h
a
n
g
e
-20
-10
0
10
20
30
0 6 12 18 24 30 36 42 48 54 60 66 72 78 84 90 96 102 108 114 120 126 132 138
Months since low
6
-
m
o
n
t
h

%

c
h
a
n
g
e
1
C
B
A
3
2 0


















-35
-25
-15
-5
5
15
25
35
D
e
c
-5
7
A
p
r-5
8
A
u
g
-5
8
D
e
c
-5
8
A
p
r-5
9
A
u
g
-5
9
D
e
c
-5
9
A
p
r-6
0
A
u
g
-6
0
D
e
c
-6
0
A
p
r-6
1
A
u
g
-6
1
D
e
c
-6
1
A
p
r-6
2
A
u
g
-6
2
D
e
c
-6
2
A
p
r-6
3
A
u
g
-6
3
D
e
c
-6
3
A
p
r-6
4
A
u
g
-6
4
D
e
c
-6
4
A
p
r-6
5
A
u
g
-6
5
D
e
c
-6
5
A
p
r-6
6
A
u
g
-6
6
D
e
c
-6
6
A
p
r-6
7
A
u
g
-6
7
D
e
c
-6
7
A
p
r-6
8
A
u
g
-6
8
D
e
c
-6
8
A
p
r-6
9
A
u
g
-6
9
D
e
c
-6
9
A
p
r-7
0
Date
6
-
m
o
n
t
h

%

c
h
a
n
g
e
-35
-25
-15
-5
5
15
25
35
0 6 12 18 24 30 36 42 48 54 60 66 72 78 84 90 96 102 108 114 120 126 132 138 144 150
Months since low
6
-
m
o
n
t
h

%

c
h
a
n
g
e
1
C
B
A
3
2 0
17% 50% 83%
100% 66% 33% 0%



















Dates
Months from cycle low
1 (High) 2 (Low) 3 (High) A (Low) B (High) C (Low)
Nov 46 Dec 57 21 29 50 81 99 133
(16%) (22%) (38%) (61%) (74%) (100%)
Nov 57 Jun 70 12 54 73 104 129 150
(8%) (36%) (49%) (69%) (86%) (100%)
Jun 70 Sep 81 10 51 69 92 123 135
(7%) (38%) (51%) (68%) (91%) (100%)
Sep 81 Nov 90 16 32 54 76 95 110
(15%) (29%) (49%) (69%) (86%) (100%)
Nov 90 Jul 02 20 46 76 95 119 140
(14%) (33%) (54%) (68%) (85%) (100%)
Averages 16 42 64 90 113 134
(12%) (31%) (48%) (67%) (85%) (100%)
Issue 45 October 2002 MARKETTECHNICIAN 17
Biases in momentum
The second type of distortion arises in relation to cycle momentum. The
distortion basically arises because of the dynamics of cycle inflexion,
although the position of the cycle within a triad may also play a role. The
point is that cycle momentum acceleratesas a cycle turns up and
deceleratesas a cycle turns down. In other words,as a cycle negotiates the
energy gap that reverses its polarity,momentum shows a massive move.
This is shown in Figure 5 below. One of the results is that the momentum
indicator will often show a lesser rate of change at the top of the cycle (ie,
at 3) than at the start of the upswing (at 1). Conversely,it will often show
a higher rate of change at the bottom of a cycle (ie,at C) than at the start
of the downswing (at B).
Figure 5:Biases in momentum
This means two things. On the one hand,a sharp acceleration or
deceleration in momentum can help to confirm that a reversal is
occurring. On the other hand,the subsequent moderation in momentum
helps to provide non-confirmation of the ensuing peak or trough. But
note that neither is valid unless it occurs in an appropriate time window.
A sharp change in momentum may occur for reasons other than a cycle
inflexion and a slowdown in momentum may be precisely what
characterises a long trend.
Market fluctuations compared
The next stage in the analysis is to overlay one market cycle on another in
order to see how closely the loci of the cycles coincide with one another.
The trick here is to ensure that the time elapse for one cycle is made
geometrically equivalent to the time elapse of another. Hence,to
compare two cycles,the time axis of one is placed on the lower horizontal
axis and the time elapse of the other is placed on the upper horizontal
axis,ensuring that the start and end of both cycles coincide. It should
then be possible to see how closely the patterns of the two cycles mirror
one another. For simplicity,I have used the 1970 to 1981 cycle as the
blueprint;but any combination could be used. Hence,Figure 6-i compares
the 1970-81 cycle with the 1946-57 cycle,Figure 6-ii compares the 1970-
81 cycle with the 1957-70 cycle,Figure 6-iii compares it with the 1981-90
cycle and Figure to 6-iv compares it with the 1990-02 cycle. Also shown
on the charts are the idealtimings of the peaks and troughs.
Figure 6-i:DJIA,1946 to 1957 and 1970 to 1981
Figure 6-ii:DJIA,1957 to 1970 and 1970 to 1981
Figure 6-iii:DJIA,1970 to 1981 and 1981 to 1990
Figure 6-iv:DJIA,1970 to 1981 and 1990 to 2002
To repeat:there are a lot of variations and distortions it would be
unreasonable to expect otherwise. Nevertheless,a quick scan of the charts
shows that (a) the threesomenessof markets comes across loud and clear
and (b) the intrinsic timing of peaks and troughs is reasonably consistent.
Hence,once we have an idea about the likely duration of the higher-level
cycle,we also have some idea about the points of inflexion of the lower-
level cycles. Cycle durations vary,but the patterns (basically) do not.
Estimating the duration of a new cycle
The problem for forecasting,however,is that we wont initially know the
ultimate duration of a newly evolving cycle. We can estimate it from the
averages of previous cycles but,as we have already seen,all we can get is
a ballpark figure. Can we be more precise? Actually the answer is,yes.
There are two ways of doing this. The first is to use the idealised
relationship between turning points. For example,once we have the first
important momentum peak,we can estimate that this peak will be about
17% of the total duration. So the estimated total duration will be:
Time from cycle low to first high/ 0.17 = First estimate of cycle length
And,obviously,the analysis is cumulative. Once the first sub-cycle low is in
(eg,after 3 to 4 years for a 11-year cycle) we can provide a new estimate:









































-25
-15
-5
5
15
25
35
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115 120
Months since low (1946-57)
6
-
m
t
h
s

%

c
h
a
n
g
e
-40
-30
-20
-10
0
10
20
30
40
50
0 6 12 18 24 30 36 42 48 54 60 66 72 78 84 90 96 102 108 114 120 126 132
Months since low (1970-81)
6
-
m
t
h
s

%

c
h
a
n
g
e
1946-1957
(l.h.scale, lower
time axis)
1970-1981
(r.h.scale, upper
time axis) 1
C
B
A
3
2 0



















-35
-25
-15
-5
5
15
25
35
0 6 12 18 24 30 36 42 48 54 60 66 72 78 84 90 96 102 108 114 120 126 132 138 144 150
Months since low (1957-70)
6
-
m
t
h
s

%

c
h
a
n
g
e
-40
-30
-20
-10
0
10
20
30
40
50
0 6 12 18 24 30 36 42 48 54 60 66 72 78 84 90 96 102 108 114 120 126 132
Months since low (1970-81)
6
-
m
t
h
s

%

c
h
a
n
g
e
1957-1970
(l.h.scale, lower
time axis)
1970-1981
(r.h.scale, upper
time axis) 1
C
B
A
3
2 0




















-35
-25
-15
-5
5
15
25
35
45
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105
Months since low (1980-90)
6
-
m
t
h
s

%

c
h
a
n
g
e
-40
-30
-20
-10
0
10
20
30
40
50
0 6 12 18 24 30 36 42 48 54 60 66 72 78 84 90 96 102 108 114 120 126 132
Months since low (1970-81)
6
-
m
t
h
s

%

c
h
a
n
g
e
1981-1990
(l.h.scale, lower
time axis)
1970-1981
(r.h.scale, upper
time axis) 1
C
B
A
3
2 0


















-20
-15
-10
-5
0
5
10
15
20
25
30
0 6 12 18 24 30 36 42 48 54 60 66 72 78 84 90 96 102 108 114 120 126 132 138
Months since low (1990-2001)
6
-
m
t
h
s

%

c
h
a
n
g
e
-40
-30
-20
-10
0
10
20
30
40
50
0 6 12 18 24 30 36 42 48 54 60 66 72 78 84 90 96 102 108 114 120 126 132
Months since low (1970-81)
6
-
m
t
h
s

%

c
h
a
n
g
e
1990-2001
(l.h.scale, lower
time axis)
1970-1981
(r.h.scale, upper
time axis) 1
C
B
A
3
2 0
MARKETTECHNICIAN Issue 45 October 2002 18
Time from cycle low to first sub-cycle low/ 0.33 =
Second estimate of cycle length
And we can average the calculations:
Estimated cycle length = (First estimate + second estimate)/2.
All that is necessary is that the highs and lows are initially approximately
in the right time frame,given momentum distortions.
Tracking a new cycle
The second way of estimating cycle duration is to track the emerging one
against its historical counterparts and infer duration from the coincidence
of turning points. However,it is important to keep a sense of perspective.
It may be reasonable to expect a significant low (say) 33% of the way into
a new higher-level cycle. However,there may well be expansions and
contractions as the cycle responds to external shocks.
5
Consequently,it is
quite important to compare the lower-level cycles directly with previous
lower-level ones. The pattern of the new cycle can then be judged a step
at a time.
An example of this is the first sub-cycle that began in November 1990. It can
be seen from Figure 6-iii that it was shorter than the 52-month sub-cycle that
began in June 1970. However,Figure 7 shows what would have happened
had we been concentrating just on the 1990-94 sub-cycle as it was
developing. By July 1994 the evidence was that it was ending. The point is
that the patterns of the 1970-74 and 1990-94 cycles were very similar.
Figure 7:Nov 90 to Jul 94 and Jun 70 to Sep 74
However,there is more:If cycles in similar positions in contiguous triads
have similar relationships to economic fluctuations,then they may also
adopt similar momentum patterns. Hence,it is a good idea to compare
the new cycle in the current triad with the analogous cycle in an earlier
triad. For example,the Dow has probably just completed the second in a
batch of three cycles that began in 1981. This was the infotech innovation
cycle. This cycle can be compared directly with the innovation cycle that
began in 1957. This was the second in a batch of three that began in
1946. It embraced the social revolution of the Swinging Sixtiesand
marked the move to mass consumption. Figure 8 below therefore shows
the locus of the Dow measured in terms of 6-month % changes
between November 1990 and July 2002 and compares it with the locus
between December 1957 and June 1970. The similarities are compelling.
Figure 8:Innovation cycles in the Dow
The current situation
Implicit in the above analysis is the assumption that the Dow completed
an 11-year cycle in July 2002 and therefore a new cycle has just begun.
6
The first thing to notice is that the 1990-2002 cycle is more complex than
some of its predecessors. In fact,the overall pattern conforms to that
shown in Figure 4. This pattern is the single most important variation of
the simple triadic pattern shown in Figure 2. This is why,as I have already
observed,it is the pattern that is the most consistent with Mr.Elliotts
Wave Principle.
The pattern in Figure 4 also seems to be the one that consistently
emerges in the context of industrial production. Unfortunately,a more
detailed analysis will have to wait for another time. However,just to give
some indication of what I mean,Figure 9 below shows the profile of the
6-month rate of change in the Dow between 1990 and 2002 against the
2-year rate of change in US industrial production between 1946 and 1980.
The patterns are essentially the same. In other words,the pattern exists in
different sectors and over different time horizons.
Figure 9:Pattern of the Dow against pattern of US output
Using the technique of comparing patterns in output over analogous time
periods,it is possible to conclude that US industrial production is now
turning up out of the infotech innovation cycle into the third (and
terminal) cycle of the current era. Figure 10 shows the 2-year rate of
change in industrial output between 1946 and 1980,and compares it with
the 2-year rate of change in output since 1980. It is clear that the pattern
since 1980 is repeating the pattern of the 34-years that preceded it. This is
consistent with a momentum low in the equity market in July 2002.
Figure 10:Comparable cycles in US industrial production
Conclusions
This analysis is only work in progress. Moreover,it is only one possible
perspective on a very complex subject. Nevertheless,the conclusions are
exciting. The very fact that an analysis such as this has been in any way
possible confirms that market price fluctuations are not randomly
generated. However,the more powerful conclusion is that the analysis
strongly suggests that market price oscillations are subject to periodic
rhythms and that these rhythms are subject to certain simple laws that
place them within the context of natural phenomena.
Very specifically,therefore,the main conclusion is that a cycle is defined,not
so much by the precision of the periodicity,as by the particulars of its pattern.



















-15
-10
-5
0
5
10
15
20
25
S
e
p
-
9
0
N
o
v
-
9
0
J
a
n
-
9
1
M
a
r
-
9
1
M
a
y
-
9
1
J
u
l-
9
1
S
e
p
-
9
1
N
o
v
-
9
1
J
a
n
-
9
2
M
a
r
-
9
2
M
a
y
-
9
2
J
u
l-
9
2
S
e
p
-
9
2
N
o
v
-
9
2
J
a
n
-
9
3
M
a
r
-
9
3
M
a
y
-
9
3
J
u
l-
9
3
S
e
p
-
9
3
N
o
v
-
9
3
J
a
n
-
9
4
M
a
r
-
9
4
M
a
y
-
9
4
6
-
m
o
n
t
h

%

c
h
a
n
g
e
-40
-30
-20
-10
0
10
20
30
J
u
n
-
7
0
A
u
g
-
7
0
O
c
t
-
7
0
D
e
c
-
7
0
F
e
b
-
7
1
A
p
r
-
7
1
J
u
n
-
7
1
A
u
g
-
7
1
O
c
t
-
7
1
D
e
c
-
7
1
F
e
b
-
7
2
A
p
r
-
7
2
J
u
n
-
7
2
A
u
g
-
7
2
O
c
t
-
7
2
D
e
c
-
7
2
F
e
b
-
7
3
A
p
r
-
7
3
J
u
n
-
7
3
A
u
g
-
7
3
O
c
t
-
7
3
D
e
c
-
7
3
F
e
b
-
7
4
A
p
r
-
7
4
J
u
n
-
7
4
A
u
g
-
7
4
O
c
t
-
7
4
D
e
c
-
7
4
6
-
m
o
n
t
h

%

c
h
a
n
g
e
Nov 90 to Jul 94
(l.h.scale, lower time
axis)
Jun 70 to Sep 74
(r.h.scale, upper time
axis)


















-20
-15
-10
-5
0
5
10
15
20
25
30
0 6 12 18 24 30 36 42 48 54 60 66 72 78 84 90 96 102 108 114 120 126 132 138
Months since low (1990-2001)
6
-
m
t
h
s

%

c
h
a
n
g
e
-30
-25
-20
-15
-10
-5
0
5
10
15
20
25
0 6 12 18 24 30 36 42 48 54 60 66 72 78 84 90 96 102 108 114 120 126 132 138 144 150
Months since low (1970-81)
6
-
m
t
h
s

%

c
h
a
n
g
e
1990-2002?
(l.h.scale, lower
time axis)
1957-1970
(r.h.scale, upper
time axis)
1
C
B
A
3
2 0





















-20
-10
0
10
20
30
40
0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 360 380 400 420
Months since low (US output, 1946-1980)
6
-
m
o
n
t
h

p
e
r
c
e
n
t
a
g
e

c
h
a
n
g
e
-15
-10
-5
0
5
10
15
20
25
30
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100105110115120125130135140
Months since low (DJIA, Sep 90 to Jul 02)
2
-
y
e
a
r

p
e
r
c
e
n
t
a
g
e

c
h
a
n
g
e
DJIA, Sep 90 to Jul 02
(smoothed, r.h.scale)
US output, 1946-1980
(l.h.scale)





















-40
-30
-20
-10
0
10
20
30
40
0 16 32 48 64 80 96 112 128 144 160 176 192 208 224 240 256 272 288 304 320 336 352 368 384 400
Months since low (1946 - 1980)
2
-
y
e
a
r

%

c
h
a
n
g
e
-10
-5
0
5
10
15
20
0 14 28 42 56 70 84 98 112 126 140 154 168 182 196 210 224 238 252 266 280 294 308 322 336 350
Months since low (1980 - date)
2
-
y
e
a
r

%

c
h
a
n
g
e
1946 to 1980
(l.h.scale, lower
time axis)
1980 to date
(r.h.scale, upper time axis)
Recovery Innovation CRISIS
Issue 45 October 2002 MARKETTECHNICIAN 19
1
See H.R. Mills,Teaching and Training. Macmillan,London,1967. It is my
opinion that Millsfindings have not been given the attention that they
deserve.
2
See,for example,Tony Plummer,Some thoughts on the mathematics of
turning points. Market Technician,March 1999,Issue No. 34.
3
This formulation appears to contradict R.N. Elliotts 5-3 pattern. In fact,it
is perfectly consistent with it. The link between the two is that Elliotts
wave pattern emerges when the market receives at least one piece of
pro-trend information during the impulse wave. It may emerge just
after the end of the re-test (information absorption) stage,or it may
occur when the impulse wave has already gained momentum. The
important criterion is that there will be an item of generally available
news to which market participants feel obliged to respond. This is a
shockto the market and,as such,will need to be absorbed. The
process of absorption will allow a significant (although mathematically
constrained) contra-trend price movement. In Elliotts schema,the
absorption of the second information shock is a wave 4. Elliotts pattern
is thus the signature of change:markets will develop five wave impulse
movements when unexpected information alters the qualitative
structure of the market.
4
Of course,this makes the heroic assumption that the July low was in
some sense the final one for the cycle that began in 1990. There are a
number of reasons for this,which will gradually become apparent.
Since the analysis is based on momentum,there remains the chance of
a further low in prices. But see also note 7.
5
In this context,monetary shocks appear especially important. In other
words,central bank monetary policy can delay or extend the lows and
highs. Ultimately,however,it cannot avoid them.
6
There is still a very big issue here,about whether or not the cycle low is
July 2002 (or thereabouts) or September/October 2001. For some time,
my preference has been for the 2001 date,mainly because the 2001
date is more consistent with developments in the US output cycle. On
this conclusion,the rally into December 2001 and the drop into July
2002 could be seen as waves 1 and 2 respectively of the (44-month)
sub-cycle that is kicking-off the new 11-year terminal cycle. It is always
possible for wave 2 to make a new low. However,it is not unknown for
the cycle low in the equity market to lag the cycle low in output. The
issue is currently unresolved. So,in order to simplify the analysis,I have
actually assumed that July 2002 is the low. In a sense,the outcome
doesnt matter for the current sub-cycle,because on either analysis the
next move is up. However,it will in due course make a difference to the
timing of the overarching 11-year cycle.
Help us to help you!
Dear STA members,
In the light of recent successes we would like to bring to your
attention the developing relationship between the STA and Robert
Walters.
Robert Walters is an international recruitment consultancy with 23
offices in 13 countries. We have extensive experience in recruiting
for some of the leading blue-chip names in finance and commerce.
We would like to take this opportunity to point out the following
aspects of our service to help further develop an effective
relationship and raise awareness about what to expect from Robert
Walters:
We take a pro-active stance in raising the STAs awareness in the
finance community. This takes time and can prove frustrating. If
you are in a decision-making position and want to hire the best
technical analysts then we want to hear from YOU!
We take pride in providing the very best in the market to clients.
This saves the client time and money but also means that not every
candidate is right for the job. Recruitment is very subjective and
for every successful placement there are many disappointed
candidates.To aid your application,our clients like to see:

Strong academics demonstrated by A levels and preferably a


business or science based degree.

Experience of finance preferably in a recognised financial


institution and not from privately run portfolios.

A broad understanding of wider issues in finance including


marketing,sales,CRM,trading.

Clear and consistent career paths with few interruptions and


jobs.

Evidence of focus and specialisation in senior candidates.


We are keen to stress that clients approach us to provide as close a
fit to their wish list as possible and that as an agency we will
struggle to provide sideways career hops without relevant
experience.
To further guide you through your job hunting,attached is a broad
salary survey:
Investment Bank Research House
0-2yrs 25 - 45K 25 - 35K
2-5yrs 40 - 70K 35 - 50K
5+yrs 65K+ 50K+
We are always happy to review CVs and try to give objective
feedback on the strengths and weaknesses of profiles. This reflects
our clients wishes and requirements for recruitment; quite often
its a story of round pegs not fitting into square holes.
As a final note the market has been tight for the last 18 months for
hiring budgets with business essential hires taking precedence.
We hope that we have experienced the worst of the downturn and
there will be an uplift in opportunities to come.
We look forward to continuing our service for the STA and clients
in the future and thank you for your continued support.
Regards,
Alec McCann
Manager
Robert Walters
0207 509 8739
alec.mccann@robertwalters.com
New element discovered
A major research institution has recently announced the discovery of
the heaviest element yet known to science.This new element has
been tentatively named Administratium.
Administratium has 1 neutron,12 assistant neutrons,75 deputy
neutrons,and 111 assistant deputy neutrons,giving it an atomic mass
of 312.These 312 particles are held together by a force called morons,
which are surrounded by vast quantities of lepton-like particles called
peons.Since Administratium has no electrons,it is inert.
However,it can be detected as it impedes every reaction with which
it comes into contact.A minute amount of Administratium causes
one reaction to take over 4 days to complete when it would normally
take less than a second.Administratium has a normal half-life of 3
years;it does not decay but instead undergoes a reorganisation,in
which a portion of the assistant neutrons and deputy neutrons and
assistant deputy neutrons exchange places.In fact,Administratiums
mass will actually increase over time,since each reorganisation causes
some morons to become neutrons forming isodopes.This
characteristic of moron-promotion leads some scientists to speculate
that Administratium is formed whenever morons reach a certain
quantity in concentration.This hypothetical quantity is referred to as
Critical Morass. You will know it when you see it...
This snippet was found on the internet and e-mailed to the editor
MARKETTECHNICIAN Issue 45 October 2002 20
I have been involved in the speculative markets for over 15 years,and one
thing I have learnt,particularly in these days of ever more complicated
and sophisticated analysis techniques,is that sometimes the simple
approach to analysis is all that is needed to be able to uncover some great
trading opportunities. One of the simplest trading patterns that I use is
the simple ABC correction.
In a previous article (Market Technician, issue No.44) I outlined how the
simple ABC correction could be used to uncover profitable trading
opportunities but, more importantly, how it was a very simple and easy
step to recognise the set-up to this pattern.In this article I would like to
take this one stage further and show how one particular occurrence of
the simple ABC correction can lead to one of the most profitable trades
available in todays markets. This is when the simple ABC correction
unfolds as part of the first correction to the first move off an
important high or low. In Elliott wave terms, this is the Wave 2 or B
correction. Once this correction is complete it can lead into a Wave 3
type move, which is usually the strongest and longest swing in a 5 wave
sequence and, as such, it carries the largest profit potential in any Elliott
wave sequence. Therefore being able to participate in this swing can
result in some very profitable trades.
The above chart shows a simple ABC correction on a UK Stock,OML.The
most important point is that this ABC correction unfolded as part of the
first correction to the initial swing off an important low.
Here you can see how OML made a major low in Dec 2001 which was
followed by an initial rally off the low. The simple ABC then appeared
during the correction to this initial rally. This is a classic Type 1 trade pattern.
So why is this pattern so important ?
Very often this initial correction is the springboard off which a very strong
move unfolds. If we move forward in time we can see how,once this
particular ABC correction was complete,OML proceeded to rally by over
30%,from a price of 89.50 to nearly 120.
A a result of the large moves that very often unfold off this set-up,
identifying these patterns can result in some very profitable trades. This is
why it is the main trade set-up I look for.
But more importantly,this set-up is very easy to identify,there is no
complicated maths or involved analysis,just a simple ABC correction that
unfolds as part of the first correction to the first move off an important
high or low.
Another example is shown in the Dow Jones chart above.Here we can see
how the Dow made an initial decline off the March highs. But then the
initial correction to this initial decline unfolded as a simple ABC
correction. This was the springboard off which a sharp decline unfolded.
This particular Type 1 trade unfolds in any market and on any time frame
(from weekly charts to 5 min charts),so whether you trade UK Shares,
US Stocks,Indices or Commodities,I do suggest that you make the
identification of this particular trade pattern one of the major parts of
your trading plan.
However,it is the simplicity of this pattern that is its strength. In todays ever
complicated and involved financial market place,where technical analysis
seems to require a degree in computer science,it is nice to see a trade set-
up that is characterised by its simplicity. This is why I believe the simple ABC
correction is one of the most overlooked and ignored patterns among the
hundreds of analysis techniques that are available today.
Steve Griffiths is the developer of the MT Predictor software program.
Steves website is at www.MTPredictor.com
Identifying latent energy in the markets
By Steve Griffiths