STATISTICAL PERFORMANCE

Hyperbolic 1 Indicator (Beta Version)

The following section will detail the statistical results produced by the Hyperbolic 1 indicator across a collection of markets and timeframes. We selected a somewhat scattered basket of markets and varied range of time frames to illustrate a wide sampling of performance statistics.  The results listed below were derived from the Atomic Trading indicator available on Bloomberg Financial.  This indicator has been available on Bloomberg for the past 2 years in varying form, as it has been upgraded over time to produce better results.

The graphic interface shown below displays a full range of selectable trading statistics (not all visible here) available for any market on any time frame that the user would like to backtest or trade.  This toolset is available as a stand-alone application through our website, without requiring a Bloomberg subscription to access and use it.  We are current developing further software partnerships to deliver this App through different interfaces, such as Market Analyst, allowing different functions to be integrated with our technology, such as automated trading, or further programming capability.

The algorithm presented in this course will produce better results than those displayed below, as some improvements have been made while reprogramming it as a standalone App. It now produces higher returns than the older version available through Bloomberg, tough the statistics below still provide an excellent example of the general functionality and output results produced by Hyperbolic 1 as programmed in our proprietary application. Hyperbolic 1-3 and Circular 1-3 will all be available through the same App.

Statistical Results - Hyperbolic 1 (Beta Version)

Across 15 Markets in Varied Time Frames (6 Equities, 3 Indexes, 3 Currencies, 2 Commodities)  

 

 

 

 

The following analysis and study of these variations is provided to help traders better understand the results of using different degrees of leverage when trading the Hyperbolic 1 algorithm.

The primary factors required to determine the viability and degree of leveraging are the number of consecutive losses, or Loss Run, and the Maximum % Drawdown variables shown in the far right columns of the statistical table on Page 102.  As will be seen below, the smaller the Maximum % Drawdown, the higher the leverage possible, and the larger the % Drawdown, the less desirable it is to leverage the system.  With this in mind, it is prudent to backtest longer data sets in order to determine probable drawdowns over extended periods.

We will give examples of three of cases below, the first showing a small maximum % drawdown (0.84%) in the USDJPY, then a mid-range drawdown (3.99%) with Google, and an extremely high drawdown (25.64%) in Facebook.  It will be seen with the Facebook example that leveraging the account above 2x wiped out the entire account, exactly the situation we most want to avoid.

The table shows a sample of 5 markets from Appendix 1, along with the results of trading them for the defined backtest period using a 2x, 5x, and 10x leverage factor.  We will illustrate the 3 cases mentioned above just to show the potential positive and negative results which can occur using these various levels of leverage with the Hyperbolic 1 algorithm. A final point is that there are other ways to take advantage of leverage besides using a margin account, such as using various options strategies which can serve to leverage returns while limiting risk to pre-defined levels.
 

 

 

Statistical Results - Hyperbolic 1 (not Beta) Indicator

Following are some of the first test statistics run on the new Excel version of the Hyperbolic 1, the first presentation of non-beta results from the new version of the Hyperbolic 1 indicator.  This first testing was done only on daily and hourly time frames, but the first results, particular for these higher time frames are excellent!

As expected the Hyperbolic produces about an average of 4-5% per month UNLEVERAGED return! What's interesting is that on the same data set in the same time frame, the Circular 1 and Hyperbolic 1 seem to produce about the same returns. However, the BIG difference is that the Hyperbolic 1 produces the SAME return with only about 1/3 the number of trades! So it saves significantly on the amount of trading effort and commission costs.  However, with further testing, this might not be the case in all markets and time frames.

Here are the results of the first 5 tests on some popular stocks using only 1 hour or daily time frames:

GE US Daily 1 Month - 11.8% Unleveraged
Leverage 3 - 36%
Leverage 5 - 61%
Leverage 10 - 127%

IBM US 60MIN 3 Months - 11.2% Unleveraged
Leverage 3 - 36.2%
Leverage 5 - 65%
Leverage 10 - 156%

CAT US 60 MIN 3 Weeks - 3.1% Unleveraged
Leverage 3 - 9.4%
Leverage 5 - 15.5%
Leverage 10 - 31%

BAC US 30 MIN 2 Weeks - 1.8% Unleveraged
Leverage 3 - 5.4%
Leverage 5 - 9 %
Leverage 10 - 18%

MSFT US Daily 3 Months - 5%
Leverage 3 - 15.7%
Leverage 5 - 26%
Leverage 10 - 53%

Statistical Results – Circular 1 Indicator

The team has also run some first performance tests on the new Excel based non-beta Circular 1 indicator. Here they randomly picked the following markets as testers, so these are not cherry-picked results looking for the very best performance.

Also, they have added a new feature to the indicator applications called Risk/Reward (RR) which allows the user to predetermine and set his risk parameters according to his comfort level. What this does is allow the user to dynamically set the stop loss level according to the desired risk exposure.  So if one were expecting a 10 point move, he could set a 10% RR and have a 1 point stop. Choosing higher risk/reward values shows that the profits increase, but this is something that the user can play with himself.

So the following statistics show the RR level (i.e. 20% = 1/5) and the total returns for that period first, at the top UNLEVERAGED. Then there are a few iterations listed below using increasing leverage on the same trades.

AAPL DAILY - 1 Year - RR 20% = 60%
AAPL leverage 3 - 180%
AAPL leverage 5 - 269%
AAPL leverage 10 - 658%

BAC 30MIN - 1.5 Months - RR 20% = 4%
BAC leverage 3 - 12%
BAC leverage 5 - 21%
BAC leverage 10 - 42%

MSFT DAILY - 4 MONTHS - RR 40% = 11%
MSFT leverage 3 - 29%
MSFT leverage 5 - 49%
MSFT leverage 10 - 101%

CAT US 60MIN - 3 MONTHS - RR 40% = 15%
CAT leverage 3 - 42%
CAT leverage 5 - 72%
CAT leverage 10 - 160%

GE DAILY - 6 Months - RR20% = 10%
GE leverage 3 - 30%
GE leverage 5 - 51%
GE leverage 10 - 103%

IBM 60MIN - 1 MONTH - RR20% = 6%
IBM leverage 3 - 21%
IBM leverage 5 - 36%
IBM leverage 10 - 77%

As you can see, the totally unleveraged returns produced by this indicator equal, at worst, the best performance of most professional managers. Add a little leverage in, and it quickly multiplies far beyond that. These systems can be diversified across multiple markets, and multiple time frames, so that they can be allocated to large scale investment and trading portfolios with 1000's of different options short, intermediate and long term options

TRADING RECORDS – CIRCULAR 3 MANUALLY APPLIED

1732% in 5 Months - Real-Time Forex Trading - Actual Account Records

Following is a set of account records demonstrating the results of actual real time trading using the Circular 3 algorithm. This is the most advanced tool presented in this series, which is specifically developed to project very accurate turning points in both price and time. 

The Circular 3 algorithm is so exact that 3 out of 10 times it hits BOTH price and time EXACTLY!  When it is not exact, another 4 of these times it is within 5% of the turning point in price and time.  The final 3 occurrences fall farther out than this and are considered misses.  Using this strategy, the Circular 3, which was NOT automated at that time, but was being calculated and traded manually, was able to produce a 1732% return in 5 months, turning a 1000 CHF account into 18324.38 CHF, a 4156.8% annualized return!!!

Looking at the statements below, on the 9th of May, just before the trade that made 3233.42 CHF there was an even 1000 CHF in the account.  Adding up all the P/L in the far right column we get 18324.38 CHF by the end of October, a 1732% profit on initial capital in 5 months.

These are the actual account records showing every trade in the sequence which produced this return. To see the types of turning points that were being identified by the Circular 3 algorithm, simply look back to the Forex charts for those currencies traded back at that time…