April 2003 Letters To The Editor
or return to April 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
HURRIED USER
Editor,
For non-US residents or just a plain hurried user, would you consider
offering a subscription to your magazine by e-mail?
Y.M.M., via e-mail
While Technical Analysis of STOCKS & COMMODITIES, The Traders' Magazine, is primarily a print publication, we do offer two online publications that may interest you. Here's an overview of those publications:
Working Money, The Investors' Magazine, offers new articles online weekly. Articles range in topics from technical analysis to the gold market to buy-and-hold investing. Our second online publication, Traders.com Advantage, offers articles daily that apply technical analysis to current market movements.
A sampling of current articles for all three publications can be seen on our homepage at Traders.com. Subscription pricing and information can be viewed by clicking on any of the three magazine logos at the top of each column.
As for STOCKS & COMMODITIES, many of the regular sections in each issue, such as Traders' Tips and website reviews, are posted online each month, as are abstracts of all S&C articles.
We hope one of our online publication options will suit your needs.--Editor
SOFTWARE REVIEWS
Editor,
I am looking for reviews of charting, trading, and analysis software,
in particular those using fractals, neural nets, and gene pattern finders,
both plug-in software and all-inclusive, standalone programs. I know I
am probably asking for an encyclopedia, but it's time for me to upgrade
from discretionary hedging to trading.
Steve Matsen, via e-mail
S&C Publisher Jack Hutson replies:
Good timing on your question -- earlier this month, all paid subscribers to Technical Analysis of STOCKS & COMMODITIES, The Traders' Magazine, began receiving copies of our 2003 Bonus Issue. The Bonus Issue contains our annual Readers' Choice Awards, which are conferred based on the results of a poll in which we asked subscribers late last year and early this year to vote for their favorite products and services.
The Bonus Issue also contains our Traders' Resource, and the category featured in that issue is software. While the Traders' Resource does not provide ratings, it does list hundreds of products related to trading and is a good place to browse for software or look up companies. You can also access this database at our website at https://technical.traders.com to browse or search by product feature.
Finally, every issue of STOCKS & COMMODITIES contains product reviews. Copies of past reviews and other articles can be purchased from the Online Store at our website, Traders.com, or found on our S&C On CD. You can also use the search window at our website to search any keyword, such as a product or company name.
PARIS TRADING
I'm sure I remember seeing two articles on "pairs trading"
in two different issues last summer in S&C. But I cannot find them.
Am I crazy or just going blind? Whatever the case, I am interested in pairs
trading as a trading strategy. Can you point to any recent articles in
S&C? I am a new professional trader and also a subscriber.
Dennis Honnold, via e-mail
In our May 2002 issue, we published "Daytrading Stock Pairs" by Mark Conway and Aaron Behle. In our March 2001 issue, we published "Pairs Trading" by Stephane Reverre.
Readers needing help locating an article published in STOCKS & COMMODITIES can use the search engine at our website, Traders.com. It can also be used to look up whether we've published an article on a particular topic.--Editor
HELP LOCATING ARTICLES
Editor,
I am a subscriber. I am looking for articles on the Hurst exponent,
or articles that have appeared in S&C on the analysis of the markets
using nonlinear statistics/indicators. Using the search feature at your
website, I found an article, but I don't know how to tell which issue it
appeared in, and therefore I do not know if it's in an issue that I have.
How does one interpret "V.14:6"? Does that mean Volume 14, issue
6?
Please let me know the issue that the article appeared in as well
as other S&C articles related to the subject.
Johannes Badenhorst, via e-mail
As for how to read "V.14:6," this does indeed mean Volume 14, Issue No. 6, which translates to the June 1996 issue. A key to which volumes were published in which year can be viewed at https://store.traders.com/articles.html.
As for the Hurst exponent, we have published several articles that touch on Hurst's methods, but not that explain the Hurst exponent. For example, see our interview with Richard Mogey (V14:7, or July 1996), who discusses Hurst's cycle methods. We also published an article by Gregory L. Morris in 1987, "Trend Of The Trend" (V5:2), which explains the Hurst method of cycle identification. Other articles may be found by using the search window at our website, Traders.com.
You might also want to try J.M. Hurst's 1970 classic book, Profit Magic Of Stock Transaction Timing (Prentice-Hall).--Editor
STEVE NISON
Editor,
Steve Nison is correct in his February 2003 Letter to S&C ("How
Reliable Are Candlesticks?"). A Japanese candle pattern is not a Japanese
candle pattern unless it is the appropriate trend -- period.
Greg Morris, via e-mail
Greg Morris is the developer of the CandlePower pattern-recognition software, which is available at MurphyMorris.com. He is also the author of Candlestick Charting Explained and has contributed several articles in the past to this magazine.
For more on candlestick charting, see our interview in this issue with Steve Nison.--Editor
EXPLORE YOUR OPTIONS
Editor,
In the January 2003 "Explore Your Options" column, Tom
Gentile of Optionetics completes his column by saying:
"With a diagonal strategy if the stock goes up, I win on my long call, and lose on my short call. If the stock goes down, I win on my short call and lose only the value of my long call."
This statement makes it look too easy. In actual fact, if the stock goes down, you only win on your short call once the cost of the long call has been covered, and the cost of that long call, because it is so far out, can be pretty hefty. If the stock goes up, it is a little easier, as the cost of the short call is far less, sometimes as much as 50% less than the cost of the long way-out-of-the-money call, by my calculations.
For example, at the time of this writing, Microsoft is trading at $53:
A July 03 40 call costs 15.60 = $1,560
A Dec 02 55 call costs 6.80 = $680 (closest to $53 on offer)
The difference is $880.
This means that the call option on the long side has to reach $2,240 -- a strike of 22.40 to break even. What is the price a Microsoft share would have to attain to reach this strike? What are the chances of it reaching this price?
Where am I going wrong?
Jacob Singer, via e-mail
Tom Gentile of Optionetics replies:
Perhaps you are not realizing that several adjustments can be made on this
position and it's not a once-in, once-out trade. Using your example:
A July 03 40 call costs 15.60 = $1,560
A Dec 02 55 call costs 6.80 = $680
(closest to $53 on offer)
The difference is $880
In order to hit maximum profit, MSFT would have to be trading at $55 per share, at the December expiration. This means the 40 calls would have at least 15 points of real value, along with any time value that may be added on, due to the fact that there is still seven months until expiration on these calls. The December 55 calls that were sold would have no value, and the spread on the trade would be at least 15 at expiration of December. Your $880 spread would grow to $1,500.
The best part about this trade is that you can now go out to January and sell calls against the Julys, and after January expires, sell February, and so on. That's the beauty of diagonal spreads. The risks are higher than a vertical spread, simply because the costs are more. Understanding how to adjust these trades as explained in the above example will significantly reduce your risk.
SHORT-TERM TRADING
Editor,
In Martin J. Pring's April 2000 S&C article, "Pick Out Your
Trading," he writes, "I'll cover these shorter-term aspects in
another article."
But since then I have not come across an article by Pring covering
this topic. Could you please ask him to write on the shorter-term aspects,
intraday trading, and so on, to enlighten readers like me?
Jagdish Chandra, via e-mail
We will pass this along. Thank you for writing.--Editor
EDGAR PETERS INTEVIEW
Editor,
I enjoyed reading your interview with Edgar Peters in the December
2002 S&C. I have a question and some comments to make.
Question: Is Peters's assessment that a deflation at this point has a low probability based on a fat-tail or normal distribution? If it is based on a normal distribution, that would contradict his statement about large events having a higher probability. If it is based on a fat-tail probability, how does this explain the fact that financial asset deflation has been so prolonged, started in year 2000 and still continuing?
Comment 1: I think Peters's view of uncertainty is a speculator's view, which most traders naturally have. But one must appreciate the fact that the effect of uncertainty is detrimental to the longer stability of an economy and especially uncertainty pertaining to the term structure of interest rates. Obviously, some, but not all, speculators thrive during uncertainty but the net effect on an economic system is highly negative. The connection he made between uncertainty and regulation was peculiar. Uncertainty is a dynamic state and regulation is a feedback, which is structured to mitigate it or, hopefully, eliminate it. The Soviet Union did not collapse because of a lack of uncertainty in the system to drive the decisions of individuals, but on the contrary, due to the high uncertainty as to its future survival in a free and global economy. I think structural uncertainty, which often benefits speculators, is often confused with dynamic uncertainty having to do with future economic conditions. The later is unavoidable in any system but the former is highly undesirable.
Comment 2: Peters is a supporter of the opinion that the efficient
market hypothesis is wrong because it assumes a consensus as to the content
of information. This statement cannot be supported logically unless one
assumes that some traders have a lower IQ and some others are highly intelligent.
My opinion is that the hypothesis is right and the reason it fails to apply
is because information has been asymmetrical. In other words, insiders
always knew more about the prospects of their companies than investors
or traders. As more controls are put in place to limit the distribution
of asymmetrical information, the markets will tend toward higher efficiency.
That is not posing a threat to trading and specifically to technical trading
as some want us to believe. Markets will trend or go sideways always, the
only difference being that profit will be strictly based on hard work in
forecasting market direction rather than "cheating" others by
being an insider and distributing asymmetrical information.
Michael Harris
TradingPatterns.com
and Harrison Investment, Inc.
ALAN FARLEY INTERVIEW
Editor,
I came up empty after reading your interview with Alan Farley ("Swing,
Swing, Swing: Alan Farley Of Hard Right Edge," S&C, February 2003).
I expected to hear something from this veteran trader about the tools and
indicators he uses to get buy and sell signals, set stop losses, and the
best practices he has formulated over the years. Maybe this info is on
his website for a price!
Richard Semock, via e-mail
NEW CHARTING STYLE
Editor,
While I was backtesting stocks, I often came up against the problem
of not knowing which of the two prices occurred first: the high or the
low. Sometimes this is very important information. To solve this problem,
I got this idea.
In candlestick charting it would be easy to draw the shadows like
this: On the left side, the extreme price that occurred first is always
above or below the body, while on the right side, the opposite is true.
Thus, this provides a little more information in the chart. Maybe you and
your readers will find this idea useful.
Wolfgang Unterwurzacher
via e-mail
NONPARAMETRIC PERFORMANCE
Editor,
I have some questions about "A Nonparametric Performance Measure,"
which appeared in the January 2003 S&C, that I hope someone can answer.
I am a subscriber to your magazine and so far am happy with my subscription.
I am a former high school math teacher with some programming skills, but I'm no Einstein. I have been plugging my way through this article. Some spots are still pretty fuzzy after about 10 readthroughs and a visit to an old stats book on the Wilcoxon signed-rank test.
First, on page 16 of the January 2003 issue, authors Guy Brys and Luc Van Hof suggest that higher ACC values should mean better indicator performance, since this means signals are being generated with relatively higher PAE values than MAE values. They show this equation:
Am I just missing something here? As the numerator of that fraction with SAE in it gets larger, ACC gets smaller, since we are subtracting the fraction from one. This numerator is larger when we have lots of positive values, which is when the signed value is positive. The signed value is positive when PAE is better than MAE. The max statement assigns SAE a value of zero when Sae is negative. So if we have a lot of zeroes, ACC will be close to one. The more zeroes (suggesting poor performance) actually gives a larger value for ACC.
Shouldn't ACC be just:
Second, the Wilcoxon signed-rank test is to be used when a dataset is possibly nonnormal. However, as I understand it, the dataset must still be symmetric.
Perhaps I don't understand the definition of symmetric, but the data shown on page 13 does not appear symmetric to me. Doesn't it appear skewed? Hence, is the Wilcoxon test valid? (My source: Milton and Arnold, Introduction To Probability And Statistics: Principles And Applications For Engineering And The Computing Sciences, McGraw-Hill, 1990, page 262.)
Third, valid system testing is new to me. Are there any reliable
books or people on the subject? For example, the authors in the "Nonparametric"
article talk about a walk-forward optimization. Just what is a walk-forward
optimization and how do I tell a good one from a bad one? When I backtest
an indicator in TradeStation, for say, 1,000 days, and I see a win rate
of 80% with a win- loss ratio of 2:1, could I be deceiving myself if I
think I have something here? What smacks of curve-fitting? Thanks for your
help.
John S. Haworth, via e-mail
Optimization is a method of fine-tuning your system to help select the
best parameter values. Walk-forward optimization is when these values are
applied to out-of-sample data in the year following the test segment. I
might suggest The Ultimate Trading Guide by Hill, Pruitt & Hill
for further information on trading systems. (Their most recent book, Building
Winning Trading Systems With TradeStation, was presented in the March
2003 Books For Traders column.)
As for your questions on the ACC formula, we have forwarded your
letter to Guy Brys for comment.--Editor
FAIR REVIEWS AND COMPARISONS
Editor,
I have subscribed to your magazine for years. However, I have been
surprised that you have never compared products head-to-head. The closest
you come is your annual Readers' Choice Awards, which in my opinion are
based largely on sales versus actual opinion. Why would someone vote for
a product they do not own? A perfect example of where you could help readers
is in comparing real-time data providers. Comparing costs, speed, reliability,
features, customer service, extras, and so on.
If the PC community can do this with all of the magazines out there, why won't S&C step up to the plate and do this? I cannot believe that with so few advertising choices, companies would pull their advertising from your publication. However, with such an obvious omission by your publication to provide real product-to-product comparisons, one can only conclude that you are concerned with your advertising revenue. So how about your impartiality to the current set of reviews?
I hope you at least have the guts to print and address this. It is funny that investors have been burned by brokers who cover the companies that pay them, and S&C, who covers trading products, appears to be doing the same thing to their readers.
I challenge you to raise the bar before one of your competitors
(albeit as few as there are) does!
Michael Schwartz, via e-mail
San Rafael, CA
We publish product reviews in every issue of STOCKS & COMMODITIES (except our Bonus Issue, which offers our Readers' Choice Awards instead). Our product reviews cover price, features, customer service, ease of use, reliability, and user opinion.
We don't usually compare products side-by-side (although we have on occasion in the past), because it can be unfair or even impossible to evaluate different kinds of products in direct comparison. Moreover, one person's opinion of a product's usefulness can be subjective and depends a lot on the trader's needs, interests, computer hardware, and Internet connection speed.
Furthermore, we have and do review data services. Already this year we have published reviews of Unfair Advantage from Commodity Systems, Inc. (January 2003 S&C), and eSignal from Interactive Data Corp. (March 2003 S&C). Readers can check whether we've reviewed a particular product by using the search engine at our website, Traders.com.
As for our Readers' Choice Awards, this feature is an opportunity for us to publish our readers' ranking of products rather than our own opinions. While this feature is to a large extent a popularity contest and has a lot to do with a product's market share (which we acknowledge in our introduction to the section), we believe it still can be a helpful tool for traders to know what other traders are using and like. While some traders have tried several products until they arrived at one they liked, other traders use the Readers' Choice ballot to simply indicate their satisfaction with a product they may be using.
Another resource we offer as a guide to products and services is the Traders' Resource. The Traders' Resource is a listing of products in 13 categories, and we print one category in each issue. This resource provides details about what the product offers, although it is not a ranking or rating. The information is provided by the product vendors via a form they fill out. The resource is browsable and searchable by feature at our website, Traders.com.
Finally, we designate most of our time, page space, and resources to articles that discuss technical trading techniques, because that is our primary editorial focus.--Editor
ERRATA: TREND AND VOLATILITY
Editor,
On page 67 of the February 2003 S&C in the article "Trend
And Volatility," there was an error in the first paragraph. Where
it says:
On August 27, 2001, MSFT was trading at 62 (Figure 5), while the September 60 at-the-money call and put were 0.8 and 5.5, respectively....
it should have read "5.5 and 0.8, respectively...."
Small error but could be confusing for an inexperienced trader.
Otherwise all great in S&C.
Mark Quantum, via e-mail
Thanks for pointing this out.--Editor
ERRATA: THE MOVING TREND
Editor,
It has been brought to my attention that there is a typo in the formula
in one of the sidebars in my article, "The Moving Trend," in
the January 2003 S&C. In the sidebar on page 38, the last formula on
the page should read:
[(6 * wsum) / (n * (n + 1))] - [(2 * sum) / n]
Please accept my apology. I have also verified the accuracy of the
other formulas. No matter how many times you reread something, mistakes
always slip through.
Bill Rafter
Futures Software, 215 829-0643
wrafter@futures-software.com
ERRATA: INGO BUCHER
The e-mail address given at the end of my March 2002 S&C article,
"Combining Fibonacci Retracements And The RSI," should have read
"iwb_ta@ yahoo.de," not "iwb_ta@yahoo.com." Please
ask any readers wishing to contact me to use this e-mail address.
Ingo W. Bucher, via e-mail
Back to April 2003 Contents