June 2003 Letters To The Editor
or return to June 2003 Contents
The editors of S&C invite readers to submit their opinions and
information on subjects relating to technical analysis and this magazine.
This column is our means of communication with our readers. Is there something
you would like to know more (or less) about? Tell us about it. Without
a source of new ideas and subjects coming from our readers, this magazine
would not exist.
Address your correspondence to: Editor, STOCKS & COMMODITIES,
4757 California Ave. SW, Seattle, WA 98116-4499, or E-mail to editor@traders.com.
All letters become the property of Technical Analysis, Inc. Letter-writers
must include their full name and address for verification. Letters may
be edited for length or clarity. The opinions expressed in this column
do not necessarily represent those of the magazine. -Editor
TRADERS' TIPS ONLINE
Editor,
Is there a way I can get the Traders' Tips from last month's magazine
online?
Steve Stoyka, via e-mail
Yes. Past Traders' Tips can be viewed at our website at Traders.com.
The current issue's Traders' Tips can be found at: http://www.traders.com/Documentation/FEEDbk_docs/TradersTips/TradersTips.html.
(Alternatively, from our homepage at Traders.com, click on the STOCKS &
COMMODITIES magazine icon on the left, then scroll down to the "This
month in S&C" section on the left-hand side, and look for "Traders'
Tips.") For Traders' Tips published earlier, click on the "Back
issues archive" located under the "This month in S&C"
section, or use the search engine. To locate tips on specific topics, use
our site's search engine.
Other code presented in the articles in this magazine can be found at
our website in a subscriber-only area at http://technical.traders.com/sub/sublogin.asp.
Login requires your last name and subscriber number.
By the way, the Traders' Tips topics this month are the RSI along with
implied volume and volatility and can be found starting on page 110.-Editor
MUTUAL FUNDS ANALYSIS
Editor,
I am a subscriber and have misplaced the S&C issue covering mutual
funds. The feature provided a listing of funds, the YTD ratio, and telephone
numbers of fund administrators. How can I review this online, or how can
I get a replacement copy?
Bernard Roy, via e-mail
The Traders' Resource topic for our December 2001 issue was mutual
funds, but you can also find the information at our website, Traders.com,
in the Traders' Resource area. There, not only can you browse the listing,
but you can also search using particular criteria based on your interests.-Editor
READER SURVEY
Editor,
I am a former subscriber and currently buy your fine publication
on the newsstand. I've been looking for the new reader survey of brokers,
data sources, and similar services, and your annual awards for the best
of those. Do you still conduct that survey? How can I get a copy of the
most recent one?
Randy Martin, via e-mail
You are referring to our Readers' Choice Awards, which are published
each year in our annual Bonus Issue. The Bonus Issue is available only
to subscribers, so that is one benefit of subscribing. Information on subscribing
is available at Traders.com, or by sending e-mail to Circ@Traders.com,
or by calling 800-Technical.
Our Readers' Choice Awards are based on a subscriber survey conducted
at our website. Results are then presented in our Bonus Issue across several
categories, including data services, brokerages, trading software, analytic
software, trading centers, and technical analysis websites. The 2003 Bonus
Issue was released in late February but is mailed to new subscribers throughout
the year. The Readers' Choice Awards are not available at our website.-Editor
SUBSCRIBER AREA AT TRADERS.COM
Editor,
In your February 2003 issue, Dennis Peterson contributed a review
of eSignal 7.1. On page 89, he mentions that the code published on that
page "can be found at Traders.com in our Subscriber Area." I've
tried looking at your website (as a subscriber to Traders.com Advantage)
but can't locate this. Can you help?
Simon Pointer, via e-mail
The Subscribers' Area can be found at http://technical.traders.com/sub/sublogin.asp.
Alternatively, go to Traders.com, click on STOCKS & COMMODITIES, then
click on "Subscribers' Area" in the contents listing on the left.
Login requires your subscriber number and last name. At this area, we post
code that has appeared in the articles in each issue of STOCKS & COMMODITIES,
as a help to subscribers when keying it in.-Editor
FINITE VOLUME ELEMENTS (FVE)
Editor,
I have read with interest Markos Katsanos' article in the April 2003
S&C, "Detecting Breakouts," on the finite volume elements
(FVE) indicator. I wondered about the following:
1. Has Katsanos tried other alternatives to the sum of the two components,
interday and intraday, respectively? One natural alternative would be to
use the either/or condition instead of the sum.
2. What is Katsanos' reason for using the 0.15% factor in the equations?
Has he tried other values?
Ake Kolm
Sweden
Markos Katsanos replies:
Thank you for your kind words about my article. In response to your questions:
1) I created the FVE indicator with the intention of combining an interday
(such as OBV) and intraday indicator (such as CMF) in one. Eliminating
either component will reduce FVE to the CMF or OBV.
2) My original threshold was zero (as with the OBV and CMF). I added
the 0.15%C (0.3% in the final formula) as an afterthought. The reason behind
it was that a stock should move at least a few cents for all the interday
or intraday volume to be allocated to the bulls or bears. I then optimized
for 0.2% to 0.7%, step 0.1%, and derived the 0.3%C in the final formula.
It is important for the coefficient to be expressed as a percentage of
the price, since a two-cent move is not significant for a $80 stock, but
it is significant for a $1 stock.
I have also tried a coefficient based on volatility. I calculated volatility
using the 22-day standard deviation and used 0.05 standard deviation instead
of 0.3% and modified the formula as follows:
PERIOD:= Input("PERIOD FOR FVE",10,80,22);
VCOEF:=Input("COEF FOR STD",0,2,.05);
VC:=Mov(VCOEF*Std(C,PERIOD),PERIOD,S);
MF:=C-(H+L)/2+Typical()-Ref(Typical(),-1);
FVE:=Sum(If(MF>VC*C/100, +V, If(MF <-VC*C/100,
-V,0)),PERIOD)/Mov(V,PERIOD,S)/PERIOD*100;
FVE
but it made little difference in practice and was not worth the complication
in the calculations. It also created some problems for a few very volatile
stocks (such as VXGN), which had to move more than 4% (which is unacceptable)
to get the volume credit.
FINITE VOLUME ELEMENTS (FVE) CODE
Editor,
I enjoyed Markos Katsanos' April 2003 article, "Detecting Breakouts."
I downloaded the coding from your website; however, I do not obtain the
same FVE chart results as those shown in the article. Are you aware of
any other readers who are experiencing this difficulty?
Steve Temple, via e-mail
See also the next letter.-Editor
CLARIFICATION: FVE indicator code
Editor,
I am a subscriber. Thanks for your recent article on the finite volume
elements (FVE) indicator by Markos Katsanos, "Detecting Breakouts."
I am in the process of programming it into Excel and noticed a possible
summation error on page 28 in the sidebar where the Excel equations are
presented.
Regarding the equations shown for columns N and O: the equation in
column N, which is assumed to be that at row 28, is given as: SUM(M6:M28).
I believe this should be SUM(M7:M28), which is the sum over the past 22
days. What is given instead is a 23-day sum (*), and works out on my spreadsheet
that I need to use SUM(M7:M28) to get close to your figure. My numbers
aren't exactly what you get because of slight differences in the volume
figures provided by my data provider. Still, could you please clarify this?
Morey Bidjarano, via e-mail
Markos Katsanos replies:
Thank you for pointing this out. The formula and the calculations are
correct, but it is confusing because row 27 should not be included in the
summations, as I inserted this later to illustrate for readers the formula
in the cells. If row 27 is deleted, then SUM(M6:M28), should be changed
to SUM(M7:M28) as you point out. I am attaching a copy of the spreadsheet
with row 27 moved to the end (Figure 1). [Editor's note: Subscribers
can visit the Subscribers' Area of our website at Traders.com to download
this Excel file.]
FIGURE 1: Here is a revised copy of the spreadsheet that was
shown in Markos Katsanos' April 2003 article, "Detecting Breakouts."
In the original, row 27 was inadvertently included in the summation process.
In this version, the original line in row 27 has been moved to the end,
and the summation is corrected to be a 22-day summation instead of a 23-day
summation. This spreadsheet can be downloaded by subscribers as an Excel
file at the S&C website, Traders.com, from the Subscribers' Area.
TRADING THE E-MINI
Editor,
Just as Willy Verwoerd recounted in his letter in the March 2003
S&C, I too was excited when I discovered the article "Trading
The E-Mini" by Dennis Meyers in the January 2003 S&C.
As my education includes extensive training in digital signal processing,
I was ready to implement the EPFFT indicator described in the article,
since terms such as "inverse FFT" weren't unknown to me. I programmed
the algorithm of the indicator in Matlab, a software tool that includes
DSP routines. Despite using e-mini S&P data of the same period as was
used in the article, my results didn't bear any similarity to the results
given in Meyers' article. I revisited the programming, but couldn't find
any inconsistencies in the algorithm between Meyers' and mine. What could
explain this?
In addition, I agree with Willy Verwoerd that this kind of article
shouldn't be used as a product review and I hope that some controls will
be implemented in the future to publish articles that are really useful.
Alejandro Fraile Vigo, Spain
Editor:
Thank you for pointing this out. Dennis Meyers' objective was to show
how advanced engineering mathematical techniques can be used to analyze
and trade noisy price series. Unfortunately, it was complicated and required
a C++ Dll to be implemented. In addition, see Dennis Meyers' own response
below:
Dennis Meyers replies:
A number of readers have complained recently in the Letters To S&C
column because no free code was given for my January 2003 S&C article
on the end point fast Fourier transform (EPFFT).
I have written more than 25 articles for STOCKS & COMMODITIES over
the past eight years. Many of my articles were improvements to public domain
systems that could be easily coded. Many were articles that used advanced
engineering mathematics to create trading systems that could not be easily
coded on today's trading platforms. To its credit - and unlike the competing
magazines - STOCKS & COMMODITIES believes its readership consists of
not only beginning traders but also advanced traders, and as such, advanced
mathematical trading techniques are considered of value to its readership.
I first presented the EPFFT in an article more than four years ago.
At that time, there was no product or code that offered to implement the
EPFFT. Due to reader demand from that article, I spent countless months
developing and debugging a C++ Dll that could be integrated into TradeStation
and MetaStock to provide the EPFFT system to anyone who wished to purchase
it. But the FFT C++ code that the EPFFT uses is not secret. It can be found
in Numerical Recipes in C++ by Press, or even for free on the web. I am
sure there are many S&C readers who have spent the time required and
coded the Epfft for themselves.
The free code concept started with S&C's Traders' Tips section.
While I commend S&C for providing this section, it must be realized
that the contributors to Traders' Tips do so for competitive reasons and
because of the free advertisement it gives them for their trading platforms.
In other words, you have to purchase the trading platform in order to use
the free code.
I would like to thank all S&C readers who have e-mailed me and told
me how much they have enjoyed my articles over the years. I hope to continue
to present S&C readers with more articles that present new trading
concepts.
DATAFEEDS
Editor,
Has there been a recent article comparing attributes (strengths and
weaknesses) of various data sources such as eSignal, RealTick, and AT Attitude,
for example?
Mike Crow
Savannah, GA
While our format doesn't include side-by-side product and service comparisons,
since we prefer to consider products on their own merits, we do occasionally
review data services, and have in the past reviewed all the ones you mention.
This year we have published reviews of Unfair Advantage from Commodity
Systems, Inc. (January 2003 S&C), eSignal from Data Broadcasting Corp.
(March 2003 S&C), and fsXtra Platinum 5.1 from FutureSource (April
2003 S&C).
Readers can check whether we've reviewed a particular product by using
the search engine at our website, Traders.com.-Editor
WHY DOESN'T THE FAT LADY SING?
Editor,
I found Bruce Faber's article "Why Doesn't The Fat Lady Sing?"
published in the April 2003 S&C very informative and enlightening.
Thank you very much for sharing that information with those of us who didn't
know about the "plunge protection team."
George Mapp, via e-mail
PATTERN RECOGNITION SOFTWARE
Editor,
I'm looking for software or a website service that can provide end-of-day
filtered searches on stocks. For example, I'd like to get a list of would-be
stocks that have just formed an NR7 day (narrowest range of the past seven
sessions). Are you familiar with any software or website that can do this
at a reasonable cost?
Leon Ho, via e-mail
You might try searching our Traders' Resource database at our website,
Traders.com, for the software features you want.- Editor
RELIABILITY OF OSCILLATORS
Editor,
We are two students writing an essay on technical analysis for our
master's degree at Stockholm University, School of Business. Our aim is
to investigate how reliable the buy and sell signals are in the oscillators
RSI, momentum, and stochastics. In brief, we will measure this by setting
up a trading strategy that we will follow. The result will be compared
with an index.
Our questions to you are: Do you know any research done on the subject
we are writing about? If not, do you know whether there are any other studies
adjoining ours? If so, where would we find it?
André Hagstedt and Fredrik Aldenstam, via e-mail
We have published many articles on RSI, momentum, and stochastics over
the years, as well as many articles on system testing. Use the search engine
at our website, Traders.com, to locate articles on these oscillators.
In addition, you may find the following articles helpful, since these
authors also set out to explore the profitability and reliability of certain
techniques or indicators, or describe how they tested a system or technique,
and what considerations they took into account during their research.
"Finding Reliable Trading Strategies" (April 1988)
The real focus of any research should be the reliability of a system during
actual trading, not the level of profits. Presented here are three methods
to help distinguish reliable from unreliable trading strategies: forward
testing, profit distribution charts, and cluster spotting. By Steve
Kille
"Profitability Of Selected Technical Indicators: Standard &
Poor's 500 Futures" (October 1987)
In previous issues, these authors reported the results of applying moving
averages, momentum, Williams' %R, Wilder's relative strength index (Rsi),
and Wilder's directional movement indicator (Dmi) to Chicago Board of Trade
corn and long-term US Treasury bond futures, Comex silver futures, and
Chicago Mercantile Exchange Imm Eurodollar futures. In this article, they
report similar information for Standard & Poor's 500 futures traded
at the International Monetary Market of the Chicago Mercantile Exchange.
By Thomas P. Drinka & Steven L. Kille
"Trading Markets With Stochastics" (December 1994)
Here's a pattern for trading markets using stochastics. By Louis M.
Lupo
"Finding Reliable Trading Strategies" (Revisited) (May 1990)
Analysts should not only identify the best trading strategy for a historical
time period but also test that optimal strategy in a subsequent time period
to evaluate its reliability in actual trading. By Thomas P. Drinka
"Selecting And Interpreting Leading Indicators" (August 1991)
Here are some techniques for selecting the most reliable indicators, interpreting
them, and checking their reliability. By Roger Pilloton
"Evolution Of A Timing Model" (February 1994)
How does a trading model evolve? Formula Research newsletter publisher
Nelson Freeburg discusses the evolution of a trading model and explains
such concepts as out-of-sample testing, parameter sensitivity testing and
the inclusion of nonprice indicators for timing trades. By Nelson Freeburg
"Basic Techniques Of Analyzing Systems" (August 1995)
What should you consider when you're designing or testing a trading system?
Here are some hints that may help you. By Mike DeAmicis-Roberts
Sidebar: Money Management
"Testing Trading Rules Over Different Time Periods" (January
1992)
In this issue's Settlement column, Technical Editor John Sweeney discusses
his process of testing and improving a system's profitability, including
the use of stops, trading windows, and selection of tradables. By John
Sweeney
ERRATA: Chicago Board Options Exchange
Editor,
I noticed on page 70 of the March 2003 issue of S&C, in the bio
of Alex Mendoza at the end of his article, the CBOE was referred to as
"The Chicago Board of Options Exchange." I have seen this error
made many times over the years, and it never ceases to annoy me. The correct
name of that exchange is the "Chicago Board Options Exchange."
Yes, its parent is the Chicago Board of Trade, but the "of" was
not part of the gift of life that the parent passed onto the child.
Mark Wolfinger, Evanston, IL
You are correct. Thank you for pointing this out.-Editor
Back to June 2003 Contents
