In this issue:
I. Fish Your Way to Successful Trading
II. MESA8 add-on review by Technical Analysis of Stocks and Commodities magazine
III. Announcing a non-linear prediction algorithm add-on from Meyers Analytics
IV. Editorial Comment by Steve Ward, CEO
V. One way to stop this newsletter
*******************************************************
I. Fish Your Way to Successful Trading
“If the wheel is an extension of the foot, then NeuroShell is an extension of the pattern-finding function of the brain,” begins Dr. Ted Achacoso, who started his trading career two years ago when he purchased the NeuroShell DayTrader. His models have become so alluring that he is now being chased by money. Friends are offering their six-to-seven figure accounts for him to trade, and a couple of them want to put together a “small” hedge fund of eight figures. “I am even receiving unsolicited checks from my family to ride on my trading account. I guess that’s the ultimate vote of confidence,” he chuckles.
Data: Ocean, Lake, River, or Stream?
“My biggest secret in creating profitable NeuroShell models is knowing the nature of the data I want to trade.” Dr. Ted likens the process to selecting the body of water where you want to fish. “While a Martha Stewart or a Ken Lay secretly can do a number on your stock overnight, countries cannot hide their GDP or unemployment numbers. That’s why I chose to trade currencies,” he explains. In addition, Dr. Ted does not trade any currency paired with the USD. “They trend,” he says, “and I am afraid of very big waves.” Rather, he trades cross-currencies, those that are not paired with the USD and move in opposite directions as a pair. His favorite is the highly volatile sterling-yen (GBP/JPY) although he does not shy away from the more docile euro-swissie (EUR/CHF). “UK’s economy cannot expand merrily while Japan’s economy contracts willy-nilly. Oscillation is inherent in the nature of these data, and therefore, repeating price patterns are more likely to appear. As a range trader, these are the waters where I fish.”
Optimizer: The NeuroShell Fishing Boat
“NeuroShell is the Rolls Royce of genetic algorithm optimizers so it will bring you to the right place where you can fish without you needing to know how it got there. You can help it along, however, by positioning it in the general area of the ocean where you want to fish. After all, it is a big ocean,” he says. Dr. Ted advocates selecting stretches of data where you can visually see or suspect repetitive patterns to be occurring. You then extend your brain’s natural pattern-finding function by letting NeuroShell extract the pattern or confirm whether or not the pattern exists. In addition, he says, “Just as you have to learn to drive a car or pilot a boat, it is better to know how to squeeze performance out of your NeuroShell optimizer rather than riding it on autopilot all the time. For example, if you need to optimize the Stochastic Slow %D indicator, set the search space from 0 to 100. There’s no sense in making the optimizer run through negative numbers.”
Indicator: Lures
“I am an indicator freak,” admits Dr. Achacoso, who has bought all of the Trader add-ons. “Each indicator is a lure that can fish for a pattern that you may not even know existed. So I think it is my duty as a trader to fill my lure box with as many relevant lures,” he says. NeuroShell is already well stocked with 800+ indicators, but he says that each new useful indicator that other traders don’t have is an edge for you. He uses the Cybernetic Analysis add-on for price pattern studies. For intermarket studies, Dr. Ted prefers the Fuzzy Pattern Recognition and Clustering add-ons. “For example, with fuzzy logic, you can say ‘buy GBP/JPY when GBP/USD is rising and USD/JPY is rising’ or with clustering, you can say ‘sell GBP/JPY when GBP/USD and USD/JPY are clustered around these values. These are true intermarket studies since they do not involve GBP/JPY at all.” Dr. Ted’s immediate excitement is Noxa’s newly released entropy indicator add-on for NeuroShell. “In less than a minute of Trader optimization, I can tell you with more than 80% certainty if tomorrow’s GBP/JPY is going to close higher or lower. How cool is that?”
Let’s Fish: Dr. Ted’s Pattern Fishing Technique
Since the Trader has so much to offer, it may appear overwhelming at first. To overcome that, Dr. Ted simply looks at NeuroShell as a three-layer cake: a data layer below, an optimizer in the middle, and indicators on top. First, he does a visual examination of the data. He identifies the most recent region of regular price oscillation and cuts this data segment on which he will build the model. Using the Trader’s prediction function, he then fishes for the price pattern within this data segment by using various indicators as inputs. His “catch” is the indicator with the highest correlation with the price, but he rejects anything below 0.8. “Kinda like throwing back fish that are too small,” he quips. Finally, he creates a trading strategy from the selected indicator. “Usually, you know you are on the right track when you fish out known oscillators, like the classic Stochastic Slow %D or the newer Ehler’s Adaptive Cycle indicator. From here, the Trader makes it simple for you to create trading strategies. If the pattern stays there for a while, you make money. The technique is adaptive because it fishes for the ‘hot’ indicator of the time segment. You don’t get stuck with constant but subjective favorite indicators,” he says. Dr. Ted uses a quadruple time frame. “The tide, the wave, the ripple, and the ripplet,” he laughs. “Range trading is notorious for open trade drawdowns, so the closer I am to my entry and exit points, the lesser my drawdown. When I tell people I am engaged in ‘risk-controlled speculation of currency’, there are three N’s for my risk control: News, Nickel – short for money management –, and NeuroShell.”
Ahoy Mate: The Ship’s Captain
“When I saw neural networks being treated as just another indicator by Trader’s optimizer, I was blown away,” says Dr. Ted Achacoso. He should know. He was a college graduate at the age of 18 and a doctor of medicine at the age of 22. Post-medically, he trained, researched, taught, and worked in three areas: minimally invasive neurosurgery, pharmacology, and medical informatics. His research, patents, software, and publications (which includes a book on biologically-based neural networks) are in the areas of biomathematical modeling of nervous systems, artificial intelligence, and distributed emergent computing. He founded and ran a group communications software company and now still provides science and technology advice to two investment funds and maintains his faculty status at the George Washington University School of Medicine.
Support: The Crew
“But you don’t have to be a physician-computational neuroscientist to use NeuroShell,” Dr. Ted insists. “Since the Trader practically drives itself, all you need is the curiosity to experiment and see where it takes you. This fishing boat is fully outfitted for deep water and stormy weather. And the crew, the technical support, is uber-responsive.” And if that is not enough? “RTFM – Read The Fine Manual! The documentation and examples contain so many precious nuggets that even experienced modelers tend to forget, “he asserts. And if that still is not enough? “Steve Ward, your CEO, called me to discuss my issues about automated trading. This reminded me of the CEO of JetBlue helping to unload baggage from his airplanes. Proof that I bought the product from the right company.”
The Final Catch
“For me, creating profitable NeuroShell models is the easy part. In less than an hour, I can teach you how to make a profitable daily model on a financial instrument that I have never seen before. My biggest problem, really, is that my brain manually overrides the signals!,” Dr. Ted confesses. “It’s funny, with NeuroShell, I fish for all of these profitable patterns and catch all of these profitable models,” Dr. Ted Achacoso concludes. “In real life, I don’t even know how to fish.”
*******************************************************
II. MESA8 add-on review by Technical Analysis of Stocks and Commodities magazine
As we announced last month, John Ehlers has completely updated his trademark MESA (Maximum Entropy Spectral Analysis) indicators in a new add-on for the NeuroShell Trader called MESA8. The MESA8 indicators were developed and programmed by Ehlers, a world renowned expert in applying scientific principles and DSP (digital signal processing) technology to the art of technical trading. The MESA8 indicators measure whether the market is in cycle or trending mode and deliver appropriate trading signals for either situation.
Now MESA8 has been reviewed by Technical Analysis for Stocks and Commodities magazine. On page 68 of the July 2006 issue, you will find Dennis Peterson’s analysis. Even if you don’t subscribe to the magazine, you can read the entire review here:
http://www.traders.com/Reprints/PDF_reprints.html$WS_MESA8
For more information on MESA8 visit:
www.neuroshell.com/addons.asp?mesa8
or
www.mesasoftware.com for more details on John Ehlers and MESA technology.
*******************************************************
III. Announcing a non-linear prediction algorithm add-on from Meyers Analytics
It is called “The nth Order Adaptive Polynomial Next Bar’s Forecast Price, Velocity & Acceleration System”. This package contains the advanced mathematical technique and noise filter called the nth Order Fixed Memory Polynomial that calculates, at each price bar, the next bar’s forecast price, velocity and acceleration. This advanced mathematical technique is currently used in today’s space-age missile and satellites applications and has been adapted by Meyers Analytics to stock and futures trading.
Using fast advanced mathematical rocket science algorithms that use discrete orthogonal polynomials, the price series is modeled using an nth order polynomial of the form:
price(t) = a0+a1*t+a2*t^2+a3*t^3+a4*t^4+…+an*t^n
Where all the coefficients are recalculated with each new price bar.
These fast and efficient algorithms use discrete orthogonal polynomials which avoid ill-conditioned matrix inversion and floating point overflow errors associated with the slow matrix inversion algorithms currently used by many to calculate the polynomial coefficients. The polynomial coefficients are computed at each new price bar and give the polynomial’s 1 bar ahead prediction for the price, velocity and acceleration. The Next Bar’s Forecast price system follows the 1 bar ahead generated curve and issues buy and sell signals based upon the curve turning up or down by a fixed percentage from a curve bottom or top. The Next Bar’s forecast Velocity and Acceleration Systems follows the 1 bar ahead generated velocity and acceleration curves and issue buy and sell signals when the next bar’s forecast velocity and acceleration cross some noise threshold.
For more information visit:
http://www.neuroshell.com/addons.asp?nth
*******************************************************
IV. Editorial Comment by Steve Ward, CEO
1. Position Management. One of the often requested features users want added to NeuroShell Trader Professional is the ability to scale or pyramid into and out of a position. In other words you start with a small position and then increase it gradually as the risk decreases, or sell it off as the risk increases. Because of the large number of requests that is clearly now one of our highest priorities for development after any lingering release 5.0 problems are totally ironed out. However, I would like to present my view of this type of “position management” and how you can already do something similar and (arguably) more effective in NeuroShell release 5.0.
The first issue that comes to mind is how do you measure risk? The time-honored way is to measure volatility – the more there is the more risky a trade might be. Then how do you measure volatility?
One way is to look at the standard deviation or the variance of prices, and both indicators are in the NeuroShell Statistical category. You can consider applying these indicators to prices, price changes, highs minus lows, or other indicators, and over varying periods of time. That’s where the NSTP optimizer becomes helpful.
There are now special indicators available that purport to measure volatility in a different way, like the CBOE’s VIX, which is even supposed to be forward looking. It works on an S&P basis instead of for individual equities. You can even trade VIX options.
Now for my idea. If a stock or other instrument is too risky to buy, then I submit that you should buy something else that is less risky, not scale a position into a risky instrument. Why not scale your nest egg into many positions in instruments based on the risk of the instrument? As more instruments become less risky, you are building a DIVERSIFIED portfolio of smaller positions in the least risky instruments, not increasing your position in only one instrument.
Let’s take an example. You’d like to buy 1000 shares of IBM, but you think IBM is too volatile now, so you buy only 200 shares, planning to buy the other 800 in increments based on some risk indicator like volatility.
In my method, you’d put the money for 200 shares of IBM into something that is NOW low risk, let’s say it is Intel. Then as time goes buy, you see than GE and AT&T have fallen into the low risk range, so you buy some increments in them. Then you see that Intel has now moved into the high risk range, so you sell off that first increment of Intel. And so on.
If you like my method of position management, you can do it now with the new release 5.0 portfolio features. Take a look at examples 18, 19, and 22 for more guidance on this. There we used the RSI indicator – just replace it with some sort of risk indicator.
2. “Will Congress keep the Web free or allow phone and cable giants to create tiers of service?” (Washington Post). I am taking about the future of the Internet, which could affect how Ward Systems and thousands of other small and medium sized businesses are built and sustained. This is a David and Goliath story, with Goliath pouring $millions into TV commercials in the US to win the battle.
The issue has been coined “net neutrality”, which is an Internet where every blogger or housewife selling hand knit socks has the same service on the Internet as huge multi-national corporations. The websites of the blogger and housewife get transmitted to the consumer at the same speed and priority as the websites of the big corporations.
But if you were the CEO of a big cable or telephone monopoly, wouldn’t you like to charge a big fat fee to those willing and able to pay more to make sure their websites had higher priority and got transferred faster than the little guy’s? Of course you would, and that is exactly what those companies want to do, according to press reports (e.g. Washington Post June 8 – “No tolls on the Internet” by Lessig and McChesney). Yet phone and cable company TV commercials make it sound like Big Brother and GOOGLE (who is on our side) want to regulate the Internet in a way that would make consumers pay more. Goliath is trying to suggest that Internet neutrality sticks it to you, when in fact, as the Washington Post describes it, “Without net neutrality, the Internet would start to look like cable TV.”
Let’s try to keep the Internet neutral so it belongs to all of us. I’ve already written to my Washington representatives.
For more info:
http://www.washingtonpost.com/wp-dyn/content/article/2006/06/07/AR2006060702108.html
http://www.google.com/help/netneutrality.html
*******************************************************
V. One way to stop this newsletter from coming
It is really easy. Just change your email address and don’t tell us.
*******************************************************