We have added EPMA (End-Point Moving Average) and Stochastics Oscillators as chartlinks to DollarLink arsenal of tools.
Last year we added the End-Point Moving Average study to DollarLink. Here is an excerpt from the January 1996 DollarLink newsletter that explains the concept.
An End-Point Moving Average is based on linear regression analysis. A regression line is a statistical concept (see the November 30, 1995 DollarLink newsletter). It is a straight line between a series of data points such that the sum of deviations of all data points from the line is the minimum. In technical analysis, the regression line typically represents a trend line that incorporates the effect of all price data. It is an effective tool because prices will tend to oscillate about the regression line. DollarLink allows you to draw regression lines and standard deviation channels above/below by pressing DDR (Data, Draw, Regression).
An EPMA is similar to an arithmetic moving average in the sense that it needs a set of datapoints (as opposed to an exponential MA which requires only the value of the current exponential MA and the new price value).
Similar to the procedure for an arithmetic moving average, DollarLink calculates the definition of a regression line for a set of data. Then it computes the value of the last (rightmost) point by using the equation of the line. That becomes the first value of the EPMA. Then DollarLink repeats the procedure for the set of data shifted to the right by one datapoint. It continues in this fashion until it goes through all the data in the chart.
The EPMA Oscillator chartlink first computes two EPMAs using the user-specified number of bars for each EPMA. It then computes the EPMA oscillator by subtracting the second EPMA from the first. (As done in all chartlink oscillator calculations, in order to bring out the difference between the two EPMAs, DollarLink multiplies the oscillator by 100.)
A Stochastics Oscillator chartlink first computes two different raw stochastics based on a user-specified number of bars. It then computes the Stochastics Oscillator by subtracting the second stochastic from the first. (Again, in order to bring out the difference between the two stochastics, DollarLink multiplies the oscillator by 100.)
Oscillators are useful indicators because they tend to accentuate differences between two different price trends. An oscillator is similar to computing the slope of a curve. In calculus (i.e. in the world of continuous -- smooth -- curves), the slope -- called the derivative in math terms -- is a useful indicator because it shows the change in direction of the curve. However, in the real world of technical analysis charting, charts are not smooth at all. An oscillator attempts to get around this by taking the difference between a fast and a slow instance of an indicator. ( Fast and slow here refer to the sensitivity of an indicator to the underlying price chart. Usually, an indicator that uses fewer bars in its computation is more sensitive -- also sometimes said to be more raw -- to changes in the underlying price chart.)
Usually oscillators use moving averages as indicators. This is how MACDs (Moving Average Convergence-Divergence) are created: first you compute the moving averages oscillator and then you compare the oscillator to the underlying price chart (by either visual inspection or, again, taking the difference between the oscillator and the price chart). You are looking for areas where the oscillator changes from negative to positive (or vice versa) and comparing that to the behavior of the price chart. Since the oscillator emphasizes trend differences, it can be used to spot trend reversals.
An EPMA Oscillator can be similarly useful. An EPMA quite often reflects the character of a price chart better than an arithmetic moving average (AMA) or exponential moving average (EMA). Thus, an EPMA oscillator should pinpoint price trend reversals better than AMA or EMA.
Raw stochastics tend to be rather sensitive indicators. A Stochastics oscillator will be even more sensitive because it emphasizes the differences of two raw stochastics. Therefore, this oscillator can be used as a leading indicator for price reversals, but care must be taken when choosing the number of price bars for the individual stochastics indicators themselves. In other words, this oscillator can become meaningless if you make the stochastics indicators themselves too sensitive.
As in all of technical analysis, there are no magical silver bullets when it comes to choosing the number of bars in a study or indicator. Each price chart has its own personality and has to be analyzed on its own terms. What works for bonds may not work for S&P, or what works for an IBM chart may not work for a Microsoft chart. Also, to make it even more challenging than that, a price chart may change its own personality as well.
Go to top | Go
To STS page | About Timing
Laboratory | Go To DollarWin
page
![[DollarLink Logo]](dlink.gif)