ANSWERS to the MOST COMMON QUESTIONS about JURIK's SOFTWARE ----------------------------------------------------------- © 2001 Jurik Research and Consulting --------------------------------------------------- | | | To DOWNLOAD, select FILE/SAVE AS ... | | | | using your browser | | | --------------------------------------------------- CONTENTS Q: WHY DO I NEED ADVANCED TECHNICAL INDICATORS ? Q: WHERE DOES JURIK RESEARCH FIT IN ? Q: WHAT QUALIFIES YOU TO DEVELOP AND OFFER THESE PRODUCTS FOR SALE ? Q: DO YOU SELL TRADING SYSTEMS ? Q: HOW LONG DOES IT TAKE TO LEARN USING YOUR INDICATORS ? Q: HOW DO PEOPLE USE YOUR SOFTWARE? Q: WHY IS YOUR SOFTWARE BETTER ? Q: WOULD I SIMPLY REPLACE MY EXISTING INDICATORS WITH YOURS ? Q: WILL THE MANUALS BE SUFFICIENT FOR A NOVICE USER ? Q. SHOULD I CONSIDER USING NEURAL NETWORKS OR GENETIC ALGORITHMS ? Q: HOW HELPFUL WOULD DEMO VERSIONS OF YOUR SOFTWARE BE ? Q: DOES YOUR SOFTWARE COME WITH A GUARANTEE ? Q: DOES YOUR RESEARCH BOOK INCLUDE YOUR MOST RECENT RESEARCH? Q: IS THERE A RETURN POLICY ON YOUR BOOKS? Q: WHAT IS YOUR POLICY ON GIVING ADDITIONAL PASSWORDS? Q: DO YOU REVEAL YOUR FORMULAS ? Q: ARE YOUR TOOLS REALLY WORTH THEIR PRICE ? -------------------------------------------------------------- Q: WHY DO I NEED ADVANCED TECHNICAL INDICATORS ? A: As with everything else in life, there are risks in trading the markets. There are also risks in buying anything to help you trade. You may have already discovered a plethora of products promising easy wealth from the push of a computer button or from pages of a book. Although anyone can randomly make a profit by shear luck, in reality, making consistent profits from the market is not a trivial task. It requires level-headed professional behavior executing a reliable trading system. Years ago, trading systems were easy to create. Books were written on how a mere moving average or stochastic could be used to signal trades. Not any more. Ever changing market behavior, made increasingly faster by computers, has made profit taking ever more challenging. Reliable trading systems now require more advanced methods of analysis to locate profit opportunities. -------------------------------------------------------- Q: WHERE DOES JURIK RESEARCH FIT IN ? A: Fortunately, with the demise of the Cold War, Wall Street took advantage of the fact that many "rocket scientists" who formerly performed arcane tasks such as analyzing signals for the military, needed a job. Wall Street has employed them by the score, and the creation of trading systems has leapt to a new level, which unfortunately is practically out of reach of the small investor. Mark Jurik is one of these former "rocket scientists", but rather than opting to work on the "street", he has chosen to form his own company. This means that the sort of powerful formulas that institutions pay dearly to develop are easily within reach of most investors and smaller fund managers. Our add-ins are designed to help you produce significantly better trading systems by "cleaning up" your financial data and, as an option, preparing your data for presentation to modeling software designed to forecast the markets. Although this may sound complicated, our add-ins are EASY to use. In TradeStation they are available as indicators and user functions. In MS Excel, you can call them from the command menu or from Visual Basic. Our user manuals provide step-by-step instructions, as well as provide demonstration trading systems. Software development at Jurik Research includes extensive testing and validation. Top quality software along with excellent manuals are our hallmark. ---------------------------------------------------- Q: WHAT QUALIFIES YOU TO DEVELOP AND OFFER THESE PRODUCTS FOR SALE ? A: I am a computer scientist with degrees in Chemistry, Electrical Engineering and Psychology. I have been developing algorithms for data analysis since 1978, mostly for the military. I went commercial as Jurik Research in 1989. I have given lectures and presentations to professional organizations for nearly 2 decades and taught classes on Artificial Intelligence for years at California Universities. My video course was approved by the California Independent Studies Program and my software tools for financial data analysis have received top scores in a review in Futures Magazine. My clients cover the globe and include large institutions, such as Merrill Lynch, Shearson Lehman, and Deutsche Bank. ----------------------------------------------------- Q: DO YOU SELL TRADING SYSTEMS ? A. Most trading systems for sale today are trivial, written by non-traders, and marketed to non-professionals. One reason for this is the task of developing professional quality trading systems is both tedious and expensive. The products are usually affordable only to large trading institutions. Consequently, we do not sell trading systems, because our price tag would be way too high for the individual investor. Instead, we offer powerful technical indicators that employ formulas similar to those found in institutional systems, as built by Wall Street's "cold-war" scientists. As for pre-built trading systems available from other vendors, most are worthless, despite their price tag. Although many vendors offer demonstration systems to prove their worth on historical data, this "test" can be misapplied and misleading. For example, we have proof that some demonstration systems distributed by other vendors were fraudulently altered to prevent trades that would have shown tremendous losses. In addition, we have seen complaints of vendors over-optimizing their demonstration system every month to produce outstanding performance on historical data, with little chance of continuing this performance on future trades. Our advice is this: if you can, create and thoroughly test your own trading system. In doing so, your trades will fit your psychological comfort zone. That is, the frequency of trades, level of risk and typical drawdowns will not stress you into the grave. For a list of good books, peruse our recommendation list at http://www.jurikres.com/faq/booklist.htm ----------------------------------------------------- Q: HOW LONG DOES IT TAKE TO LEARN USING YOUR INDICATORS ? A: It is often said that a guitar is the easiest instrument to learn, yet you can spend years mastering all its potential. Likewise, our modules require only a few minutes to figure out how thew work, yet our customers comment that after experimenting with the tools for a while they have discovered amazing applications not possible with classical technical indicators. ----------------------------------------------------- Q: HOW DO PEOPLE USE YOUR SOFTWARE? A. **** VEL **** Users typically try the demonstrations in our manuals and then experiment. For example, one customer discovered VEL can pick price runs driven artificially by floor traders. He identifies moments when floor traders jack up prices unreasonably quickly to clear out their books. So when VEL rises above a very high threshold, the odds are you can expect a reversal. Another investor used to estimate the market's momentum using the slope of his regression lines. He later discovered that VEL produces superior momentum estimations and removed all regression calculations from his trading system. **** CFB **** The primary use for CFB is to estimate how long in time a price time series was trending and is not an overbought/oversold oscillator. Users have used this estimate to determine the proper length of classical indicators. For example, one customer found good results when RSI was made longer as trend duration became longer. CFB was essential to this strategy. If a market fails to reverse and resumes a trend, a simple way to protect yourself against an unexpected breakout is to detect them with adaptive channels. The CFB user manual shows one example of how CFB can be used to dynamically adjust a breakout channel for optimal performance. The system's formula is also provided, serving as a framework for your own experiments in adaptive channels. Another investor uses CFB to gauge how much retracement is needed before entering a trade. The idea here is that he wants to enter on a reversal when the reversal itself is of a size proportional to the trend that's just about to end. That is, the larger the trend that just died, the greater the required reversal before entering the market. **** AMA and JMA **** The primary purpose of AMA and JMA is to reduce the noisy (jumpy) action of any time series while adding as little lag as possible. Lag avoidance is critical as late trades are typically less profitable. Both tools can detect and respond to price gaps and rapid trends better than any other moving average or filter. Because of this, both tools have near limitless applications. A classic application is to create an improved MACD using either tool. You'll discover the crossovers are more timely with far fewer false signals than a classical MACD. Both tools can filter other indicators as well as market data. **** COMBINATION **** A sophisticated approach for intraday S&P is to identify major swing points. A method is to use AMA (or JMA) to smooth a fast %K to determine when the market is trading in the extremes of its trading range (>90%, <10%). Next, when a trend loses steam, CFB will likely fall rapidly from a large value. So when both conditions are met, you may want to take a position favoring a reversal. Finally, when the market reaches higher highs but VEL produces lower highs, this divergence is a strong indicator of impending reversal. Using all 3 tools together, you can produce a very strong swing indicator. I've been notified of many discoveries by our customers. It is safe to assume even more methods are being kept secret. ---------------------------------------------------- Q: WHY IS YOUR SOFTWARE BETTER ? A: Two reasons: better algorithms and better product development. One ingredient to winning the markets is to remove as much lag in your timing as possible. You simply cannot compete against traders pre-empting your trades by several bars. They will get the lion's share of profits and you get the scraps. Standard technical indicators have lots of lag, the primary cause for late trades and lost profit potential. Creating indicators that are timely, responsive and smooth is a very difficult task. Consequently, indicators available from vendors also suffer from excessive lag. Jurik Research algorithms are specifically designed to provide meaningful technical indicators with AS LITTLE LAG AS POSSIBLE. This helps you make earlier trades, giving you and your account a bigger piece of the action. As for product development, AMA, VEL, CFB, DDR, WAV and JMA modules are first designed and tested using MATLAB, the engineer's choice for heavy duty signal processing. We then code and test algorithms for TradeStation, MS Excel, and we also produce DLL modules for program developers. All versions undergo rigorous in-house alpha testing and external beta-testing before being released to the public. In addition to our rigorous testing, we supply up-to-date informative user manuals. We immediately apply user feedback to improve their clarity and enhance their utility to the user. We strive to assure product quality, and our satisfied customers say we are achieving that goal. You can view their letters at ... http://www.jurikres.com/quotes/mainquot.htm --------------------------------------------------------- Q: WOULD I SIMPLY REPLACE MY EXISTING INDICATORS WITH YOURS ? A: In some cases, yes, but in others, no. For instance, you might well simply replace "momentum" with our VEL. However, our AMA has so little lag, it makes you rethink how lag plays a role in technical indicators. For instance, the reason MACD works is the lag BETWEEN the fast and slow moving average signals. Knowing this, you can improve the MACD by using AMA to product the faster line and a simple moving average to produce the slower line. This simultaneously maximizes the lag between the two lines as well as minimizes the delay in attaining crossover signals. Its surprising how many customers discover this on their own. In short, our tools help you control lag and provide new opportunities to improving your trading systems. --------------------------------------------------------- Q: WILL THE MANUALS BE SUFFICIENT FOR A NOVICE USER ? With our tools, there are no mathematics you need to calculate. You simply set parameters as you would with classical indicators, (moving average length, for example). Software developers wanting to embed Jurik's generic DLL files in their own software applications will need to write code to access them. Our user manuals include background material about the indicator, and one or more demonstration examples for you to try out. Jurik's tools for the TradeStation platform also include a one page "executive summary", designed for those who don't like reading user manuals and prefer to have all pertinent information on one page. Jurik's generic DLL files for software developers includes sample code showing how to call it in Microsoft C. Are you tired of user manuals that have just too many errors. Our policy is to examine and revise manuals each time an error is brought to our attention. This way, when you purchase our software, the manual is very accurate. In fact, it is also our policy that if discover an error in any of our manuals, we'll credit you $50 off any future purchase of our software tools, including upgrades. No other vendor stresses quality as does Jurik Research. --------------------------------------------------------- Q. SHOULD I CONSIDER USING NEURAL NETWORKS OR GENETIC ALGORITHMS ? A. Both Neural nets and genetic algorithms are for advanced traders looking to push their trading system to higher performance levels. Neural networks are useful for developing leading indicators, such as forecasting the highest high and lowest low of the next 8 bars. A good neural net product will automatically try various combinations of inputs to get the most reliable performance. BrainCel develops neural nets in MS Excel spreadsheets, and the forecasting models produced can be used in both MS Excel and TradeStation. This lets you build real-time trading systems that employ neural net based indicators. Genetic Algorithms (GA) are useful for developing and fine-tuning trading rules. Evolver is a GA that works in MS Excel spreadsheets, whereby you would place your indicators on the spreadsheet, and design trading rules that have adjustable parameters. These rules could, for example, select which stocks or markets may be traded that day as well which indicators to use, and what patterns to look for within the indicators. You then let Evolver adjust these parameters in order to optimize trading performance. --------------------------------------------------------- Q: HOW HELPFUL WOULD DEMO VERSIONS OF YOUR SOFTWARE BE ? A: Not very. That may come as a surprise, but there is sound logic behind it. Standard technical indicators have significant lag. For years users have grown accustomed to it and tried to devise ways to mitigate its disadvantages. Because our tools have little lag, proper application requires spending time rethinking how to write technical indicators THE WAY THEY SHOULD HAVE BEEN ALL ALONG. For example, a common first-approach toward creating a MACD with JMA is to use it for both the fast and slow crossover lines as follows: MyMACD = JRC.JMA(close,6,0) - JRC.JMA(close,30,0); However, the result is mediocre. This is because MACD actually *requires* lag, but only *between* the two crossing lines. Therefore, for best results, it's better that the slower line have as both lag and momentum so that it does not weaken when price reverses. Otherwise, the crossover is delayed as the faster line tries to meet the slower one (which is turning away). A simple moving average has more lag and momentum than JMA. So a better MACD would be as follows: MyMACD = JRC.JMA(close,6,0) - average(close,30); Learning when and where to add or remove lag in technical indicators takes time, and we found that during a brief trial period, the user is not committed to making the effort, and thus may decide JMA is not much better than what he already has. A few demonstration charts beyond those displayed on our web site would not offer added value because they lack the interaction needed to realize their true power. In contrast, customers who buy our tools dedicate time to go through our manuals and inevitably discover new ways to make indicators. Eventually, they decide our tools are excellent because they open new possibilities for technical analysis. The varieties of ways users have applied CFB, for example, amaze me. We realize that having no free trials will discourage some from acquiring our software, and that's OK. To help in the decision making process, our web site displays numerous charts that point out JMA's advantages, and posts a large collection of letters from users. We also try to provide quality tech support (within reason). All in all, we're happy to have a user satisfaction rate better than 99.5% --------------------------------------------------------- Q: DOES YOUR SOFTWARE COME WITH A GUARANTEE ? A: It is very rare that any vendor selling technical indicators will include a guarantee, BUT WE DO! We obviously cannot guarantee that you will become filthy rich, simply because although we provide you with superior tools, you still need to create your own trading system. What we do guarantee is that OUR TOOLS PERFORM AS ADVERTISED. If during the first month of your purchase of our modules, you can prove they do not perform as advertised, WE WILL REFUND YOUR MONEY. We have that much faith in our products! Regarding JMA ... The purpose of any moving average is to extract a smooth signal from a noisy one. We cannot guarantee that JMA will improve ANY arbitrary trading system as that would be an impossible task for any moving average. What we do guarantee is that our JMA will smooth your data with the best combination of all these desirable properties: 1. low lag 2. low approximation error, 3. low overshoot and undershoot, 3. superb smoothness If, after acquiring JMA, you can provide us with an algorithm that does a better job at all the above four tasks simultaneously, we will refund your money. To date, after seeing JMA in action, no one has found a better algorithm. See the FAQ section of our web site for additional details on the JMA guarantee. Or simply go to ... http://www.jurikres.com/faq/faq_ama.htm#guarantee ------------------------------------------------------------ Q: DOES YOUR RESEARCH BOOK INCLUDE YOUR MOST RECENT RESEARCH? Q: IS THERE A RETURN POLICY ON YOUR BOOKS? No. Research is a very expensive process. Purchasing the book gives the customer access to our past research. We make no representation that our past research will satisfy the customer's current needs. The book does not include our most recent research. Access to recent works requires hiring us as consultants. --------------------------------------------------------- Q: WHAT IS YOUR POLICY ON GIVING ADDITIONAL PASSWORDS? We understand that some users want to run Jurik software on a computer at home and another at work. Also, a new password may be required when the user replaces an old computer with a new one. To accomodate these situations, we have the following password policy.... Jurik Research provides additional passwords to enable installation onto additional computers owned by the licensed user. The first additional password is free, and there is a $100 fee each one beyond that. However, that fee is waived if the licensed user can document he is replacing an old computer that had Jurik's software installed, with a new computer. A faxed copy of the new computer's invoice would suffice as proof. --------------------------------------------------------- Q: DO YOU REVEAL YOUR FORMULAS ? A: Our formulas represent years of privately funded research, amounting to a considerable expense. In order to preserve our "intellectual property" we must protect the algorithms from competing vendors. They would love to know how our indicators work. Parts of our algorithms are registered with the U.S. Copyright Office and parts are trade secret. No part of our algorithms is publicly disclosed. This combination gives us the most leverage in warding off illegal acts of piracy. We have successfully litigated against every violator brought to our attention. --------------------------------------------------------- Q: ARE YOUR TOOLS REALLY WORTH THEIR PRICE ? Our success depends upon the success and satisfaction of our customers, and we have lots of satisfied customers. Many have said our tools were one of their best investments. Check out their letters at http://www.jurikres.com/quotes/mainquot.htm . Our prices are reasonable. Our customers are devoted. Our software is guaranteed. The next move is yours. Regards, Mark Jurik ******************************************** Tools for financial data preprocessing Jurik Research http://www.jurikres.com/ ********************************************