|
ADSI includes cutting-edge techniques based on digital signal concepts. As a group, these tools minimize or eliminate lag, rapidly identify cycles & trends, and measure statistical extremes.
ADSI's techniques are based on Dr. John Ehlers' research into applying digital signal theory & cybernetic analysis to markets. Although the theory is complex, the application is not. These tools provide clear, responsive measurements of market conditions. They are powerful, easy to use, and process quickly in real-time.
ADSI's techniques are modular. Wherever possible, adaptive "building blocks" of Dr. Ehlers' indicators have been provided. He encourages creative application of his concepts. ADSI's modular approach allows users to construct their own methods based on his research.
In many cases, users may select from several algorithms to provide the variable period these indicators require as input. In others, the theory is supported only by the use of a specific algorithm. Refer to the ADSI Technical Documentation page for guidance on this matter.
The indicators integrate into Metastock as external formulas. They are flexible and easy-to-use like traditional Metastock indicators, and process quickly as C++ dlls. Advanced design has been used to optimize their computational performance. They are ideal for use with real-time or end of day data. Refer to the Metastock user manual for information on plotting external formulas.
The indicators included in this section are listed below. Reference material which describes their development & interpretation is also noted. The number(s) in parentheses shown after the indicator refers to the chapter in the referenced book. If two chapters are referenced, the indicator has been derived from combining concepts in both chapters. If an "x" is listed after the chapter number, the indicator has been derived by combining concepts in the chapter with those in the traditional formulation of the indicator.
- Enhanced Signal to Noise Ratio (8)
A responsive method to identify noisy markets.
- Hilbert Oscillator (8)
A responsive oscillator based on the Hilbert transform.
- Homodyne Cycle Period (7)
A low lag discriminator for measuring cycle period lengths.
- MAMA & FAMA (17)
A combination of nonlinear adaptive moving averages designed to minimize whipsaws. A cycle's phase rate of change governs their reponse to market conditions, and whether MAMA crosses FAMA or not.
- Optimum Predictor (20)
A predictive oscillator applied when markets are in a cycle mode.
|
- Hilbert Transform (9)
A technique to identify components required for cycle measurement. This version includes enhancements to minimize lag.
- Instantaneous Trend (3)
A zero lag average of price.
- Cybernetic Cycle Content (4)
The portion of price related to a cycle. A simple price model is: price = trend + cycle. If the price and trend value are known, the cycle content can be derived. The Instantaneous Trend is a dual of this indicator.
- Cybernetic Cycle Period (9)
The cycle content and a median filter are used to measure the cycle's period (or "length"). This technique is much faster than traditional cycle discriminators including: fast fourier transforms, homodyne discriminators, and even maximum entropy spectral analysis (mesa). The cycle period may be used as input to make traditional or custom indicators responsive.
- Adaptive Momentum (12)
The cycle period is used to create a responsive, adaptive measure of momentum.
- Adaptive Cyber Cycle (10)
The cycle content and cycle period are used to create a responsive, adaptive oscillator.
- Adaptive CG Oscillator (10)
The standard center of gravity oscillator and cycle period are used to create a responsive, adaptive oscillator.
- Adaptive RVI (10)
The standard relative vigor oscillator and cycle period are used to create a responsive, adaptive oscillator.
- Sinewave Indicator (11)
A unique approach to rapidly identify trends,
and cyclical turning points. In Cybernetic Analysis,
Dr. Ehlers built on his earlier research into
this technique to create an even more responsive
version.
- Two Pole Smoothers (13)
Adaptive, highly efficient data smoothers. There are two versions:
- Two Pole Butterworth Filter
- Two Pole Super Smoother
- Three Pole Smoothers (13)
Adaptive, highly efficient data smoothers for additional smoothing. There are two versions:
- Three Pole Butterworth Filters
- Three Pole Super Smoothers
Stochastic-Fisherized Adaptive Indicators
Extensive research indicates market returns are not normally distributed, yet many analysts apply techniques which assume they are. The Fisher transform converts standardized data sets to an approximately normal distribution. Statistical tests based on the normal distribution may then be applied. Data sets must be range bound between -1 and +1 before the Fisher transform may be applied. The "Stochastic" standardization technique is used to accomplish this.
Several applications of the Stochastic-Fisherized approach are provided. Each creates adaptive oscillators with sharp, statistical measurements of price extremes. The applications provided are as follows:
- Stochastic-Fisherized Adaptive Indicator (1 & 10)
The Stochastic-Fisherized technique is applied to a user selected data array or indicator. As a result, a user may create custom-designed, statistical oscillators.
- Stochastic-Fisherized Adaptive Cyber Cycle (1 & 10)
The Stochastic-Fisherized technique is applied to the Adaptive Cyber Cycle.
- Stochastic-Fisherized Adaptive CG Oscillator (1 & 10)
The Stochastic - Fisherized technique is applied to the Adaptive CG Oscillator.
- Stochastic - Fisherized Adaptive RVI (1 & 10)
The Stochastic-Fisherized technique is applied to the Adaptive RVI.
Laguerre Filters (14)
A special class of filters which "warps" short data arrays to provide extremely responsive, but very smooth indicators. The Laguerre Filter may be applied to create averages, oscillators, or other indicators. Several applications of the Laguerre Filter are provided. Each creates powerful indicators with very little data (three to eight samples). The applications provided are as follows:
- Generalized Laguerre Filter (14)
The Laguerre Filter is applied to smooth price or indicator data over a user specified interval, typically four to seven periods.
- Laguerre CCI (14x)
The Laguerre Filter is applied to price data to create input for computing the Commodity Channel Index (CCI). The CCI is computed using this result.
- Laguerre RSI (14)
The Laguerre Filter is applied to price data to create input for computing the Relative Strength Index (RSI). The RSI is computed using this result.
- Laguerre Stochastics (14x)
The Laguerre Filter is applied to price data to create input for computing Stochastics. Stochastics is computed using this result.
|
|
|
