Using the 2nd Derivative of Asset Price to forecast trend changes
Harnessing Market Trends – Limitations of Trend-Following systems: Markets tend to trend, as is explained in Why Absolute Return?, and our goal at THE ABSOLUTE RETURN is to harness these trends into investment returns. There are a plethora of trend capture methods that have been discovered and defined in the field of Technical Analysis but a large majority of these fall into the category of trend-following systems. In general, trend-following systems tend to be mechanical, rule-based systems which succeed in strongly trending markets, but suffer from two draw-backs -
-
Trend-following systems typically sacrifice a percentage of gains near the tops and bottoms of an asset price and capture a portion of the gain in the middle, and
-
Trend-following systems tend to perform in strongly trending markets, but returns are poor to negative in sideways markets because such markets typically result in many false trend signals.
Sometimes, such sideways markets can go on for a while and result in negative returns or higher draw-downs which translates to higher risk. It takes considerable skill and a very good feel for the markets for a money manager who employs trend-following technical trading systems to apply the right system to the appropriate market environment, or manages positions wisely in an environment which is adverse to the trend-following system being employed. This is exactly what we try to do at THE ABSOLUTE RETURN in our Market Neutral Long/Short portfolio.
Precision Timing of Highs and Lows – The holy grail of investing: But is there a way to time the tops and bottoms inevitably present in an asset price? And is there a way to predict whether the next trend is more likely to be a strong upward one, a strong downward one, a sideways winding road, or something in the middle? This is the “holy grail”, so to speak, of market-timing and broadly speaking of investing. The experts deny this is possible and the many of the best market timers state that they are satisfied with capturing the “middles” of trends. The Elliott Wave Principle, and our relatively more objective application of it at THE ABSOLUTE RETURN, is a potential solution, and indeed there have been some remarkably accurate forecasts of major price tops and bottoms made by the Wave Principle, but we have found that it is not reliable enough by itself. We have discovered and developed a proprietary method that works synergistically and symbiotically with the Wave Principle that takes significant strides towards this ideal. This method is based on a study of the 2nd derivative of the price of an asset.
Principles found in a study of the 2nd Derivative of Price: The best way to explain the notion of trying to predict the movement of price over time, particularly the points at which the price of an asset tops or bottoms, is to use the analogy of a ball thrown into the air on a wind-less day. For a ball thrown into the air, in a plane perpendicular to the ground, where the information available is a plot of the initial set of positions of the ball with time, it is possible to predict the remaining trajectory of a ball. This is because this information can be used to derive the key variables which are the initial momentum vector (mass times velocity) and the drag force against the ball which is a result of air resistance and the roughness and geometry of the ball. Hence the point at which the ball reverses direction (i.e. the top of the trajectory), the point at which it hits the ground (let us call that the bottom of the trajectory), and even the time taken for the ball from its launch to when it hits the ground, are predetermined by the initial momentum vector and initial plot of ball position with time (with as few as 3 data points on the parabola). In a similar fashion, simplistically speaking, we can gain some very useful information from the initial “trajectory” of an asset price when it begins a trend, or from how a price proceeds in a trend, to “derive” the likely price level at which the asset could change its price trajectory, i.e. where the price of an asset could top or bottom. It is actually not as simple since there are a plethora of variables (time-varying forces and the emergence of new forces that manifest themselves during a trend) involved which defy derivation, but a careful study of the 2nd derivative of Asset Price yields surprising treasures. I will keep the following discussion general on purpose to protect the proprietary nature of the discoveries and applications, i.e. protect the “secret sauce”, but here is the gist of what we have found.
The 2nd derivative of the position of a ball from the above example is a constant. The 2nd derivative of price on the other hand is not a constant (market timing would be too easy if that was the case), but exhibits regularities which when analyzed from within the framework of the Elliott Wave Fractal model of prices, yields the following information.
i) It gives a clear and objective “early-warning” of a coming change in price trend at the degree of interest of the trend, i.e. secular, primary, intermediate, short-term, and even intra-day trends.
ii) It provides some very important clues on the nature of the next price trend, i.e. whether it would be a strong trend or more of a sideways type trend.
iii) It confirms when the price trend has changed, typically sooner than most trend-following systems.
iv) It can provide important clues on the longevity of the next trend.
In summary, our research indicates that the prior price trajectory of an asset, when analyzed in the context of the Elliott Wave Fractal Model, pre-determines and constrains some key aspects of future price behavior of the asset. These conclusions can provide a very important clue in our quest to time the Top or Bottom of an asset price as close to the actual top or bottom as possible, and as reliably as is possible, within the inherent uncertainties of the real-world and the limitations of human analysts.
Back-testing results: Two of many examples of the results of back-testing this model have been provided below, one at the secular trend timeframe for the Dow Jones Industrial Average (DJIA) and one at an intermediate-term trend timeframe for Crude Oil. For each top and bottom at the desired degree, three-signals are listed – the pre-warning or trigger, the actual buy or sell-short signal, and the confirmation signal for the new trend. The buy signals are shown with Blue (pre-warning) and Green (confirmation) rectangles with Green arrows (actual model Buy signal) which signal a major low. The sell signals are shown with Yellow (pre-warning) and Red (confirmation) rectangles on the price chart and Red arrows ( actual model Sell signal) which signal a major high.
1. Dow Jones Industrial Average (DJIA) Secular Trend Forecasting/Identification from 1905 thru 2010:
