October 1996
Letters to the Editor
Editor,
GREAT MINDS THINK ALIKE
I'm so glad I subscribe to STOCKS & COMMODITIES. What great reading from great minds! I would like to encourage Tushar Chande to write more articles describing his ideas. I enjoyed his past articles in S&C and I would like to read more. I also enjoy any article that deals with trading longer-term positions, such as weekly charts.EUGENE T. McENERY III
via E-mail
TESTING AND HARD WORK:
Editor,
KEYS TO SUCCESS
Technical analysis and trading systems are often considered pseudo-sciences. One reason for this reputation is the widespread use of misapplied scientific methods such as curve-fitting and the use of anecdotal examples as proof. Another reason is that technical analysts sometimes act as though they're members of a secret society or modern-day guild, who believe they have special technical analysis secrets that only they know, and that allows them to garner abnormally high profits from the markets. The truth is, most of the so-called secrets, or trading ideas and techniques, are already in the public domain. What's difficult is determining which of the trading systems in the public domain work and which don't. This work requires testing the various systems and then analyzing the results on out-of-sample data.I believe it's possible that someone could give away a great system for free, and yet the lemming crowd -- those traders who hunt for the Holy Grail instead of working, testing and perfecting the ideas they already have -- would completely ignore it, not placing any value in a system that didn't cost a large sum of money. In my opinion, many trading systems that have been presented in STOCKS & COMMODITIES can give as good results as any high-priced system you can buy. In a secret guild, the obvious is dismissed because it's not secret and thus is deemed valueless. The "secret guild" mentality perpetuates the myth that if something is free or obvious, it can't possibly be any good.
DENNIS MEYERSSee Dennis Meyers's "The siren call of optimized trading systems" in this issue. In this article as well as in his past articles, he presents the results of rigorous system testing. -- Editor
Chicago, IL
meyersx@enteract.com
ADS FASCINATE TOO
Editor,
I was fortunate to discover a copy of your magazine in a second-hand book store. I am fascinated by the articles and the advertisements the magazine carries.I would be grateful if you could furnish the address for the software company Omega Research, which produces the program SuperCharts. It would be better still if you could forward my request for details regarding cost, availability, features and so on to Omega Research.
B. THEER THES WARAN
Madras, IndiaOmega Research is located at 9200 Sunset Drive, Miami, FL 33173, phone 305 270-1095 or 800 556-2022. -- Editor
LINEAR REGRESSION AND T-BOND YIELDS
Editor,
I have a question about "Treasury bond yields and the S&P 500" in the August 1996 STOCKS & COMMODITIES. The linear regression formula was presented as:Monthly % change in the S&P 500 = 0.859% - 5.947 * (yield - moving average)I believe the 0.859% is the monthly S&P return, but I'm not sure what the 5.947 represents.
DAVE DORGAN
via E-mailThe formula for a linear regression line is Y = bX + a, where b is the slope of the line; that is to say, if X increases by one unit, we want to know how much Y changes. The a is referred to as the alpha and represents the value of Y if X = 0. For example, if the difference between the current T-bond yield and the six-month moving average of the T-bond yield is zero, then the stock market has tended to rise by 0.859%. If the T-bond yield is 1% above the six-month moving average, then the stock market has tended to fall by 5.088%, because 0.859% - 5.947*(1%) = 5.088%.
-- Editor
SEQUENTIAL CANDLESTICKS
Editor,
I read an article about using Tom DeMark's Sequential system with candlesticks, but I'm not sure if it was in your magazine or not. I was wondering if you could tell me if it was and what back issue it appeared in so I can order it. Thank you.K. KIELY
via E-mailThe March 1995 issue of STOCKS & COMMODITIES (Volume 13) included "A trade in foreign currencies" by D.W. Davies, which incorporated the Sequential system with candlestick charting.
In addition, we interviewed Thomas DeMark in the May 1995 issue of STOCKS & COMMODITIES.
Contact our circulation department at 800 832-4642 or https://www. traders.com/ for ordering information on that issue or the 1995 Volume book compilation. -- Editor
CALCULATING ALPHA
Editor,
In the article "An alpha indicator for bonds" (STOCKS & COMMODITIES, July 1996), an Excel spreadsheet formula is given for calculating alpha. Since I don't have Excel, I tried calculating alpha using the formula for calculating the least squares line obtained from the book Statistics by Chester L. Olson. I cannot duplicate the answers given in the sidebar. The formula I am using is given as:where:
These formulas are given with the explanation, "Differential calculus can be used to show that values of b and a that minimize the least-squares criterion." Can you help me?
E. DONALD WILSON
Calgary, Alberta, CanadaWe checked your formulas in Excel, and they are correct. Perhaps you reversed your x and y columns. -- Editor
POSITIVE PRODUCT REVIEWS
Editor,
Your magazine is really good, but there's one area that could be improved: It seems that all the software packages reviewed by your magazine are recommended as good products despite their shortcomings. Some of these reviews are fluff pieces; it seems like you don't really want to be objective for fear of offending someone.MARK
via E-mailWe receive many more products for review than we have time or space to publish. We consider it a waste of editorial space to publish reviews of poor products. Rather, we try to only publish articles and reviews that will help our readers trade profitably.
Thanks for your comments. We are always trying to improve the magazine. To improve, we need feedback. -- Editor
CONTINUOUS FUTURES CONTRACTS
Editor,
Good job on putting together a nice Web page.I need to acquire a copy of the article that you published several years ago on how to construct a continuous futures contract. If you can, please mail me a copy of that article.
CHARLES ARTHURThe following article may provide the information you are looking for:
Telluride, CO"Selecting the best futures price series for computer testing"
(S&C, October 1992)One problem that traders studying commodity markets face is the fact that individual futures contracts have price characteristics that are not continuous with other contracts within the same market. Jack Schwager, author and director of futures research at Prudential Securities, has some suggestions on dealing with this problem.
By Jack SchwagerWe do not mail copies of articles. Single back issues for the current and recent years are available for $8. Alternatively, you may purchase the annual book compilation for the year in which the article in questions appeared. To place orders, please visit our Web site at https://www.traders.com/ or call us at 800 832-4642.
ERRATA
Editor,
What am I missing? The formula in Traders' Tips for the MetaStock version of the trading system described in Dennis Meyers's article "The gold and silver...." seems to be incomplete for the "Enter long" formula. Can you help? Thanks.BOBIt looks like there is indeed something missing. My recommendation is the numerator of the enter long should be:
via E-mail
-- Editor100*(P-Ref(P,-Opt1{jmplag}))/
