SEPT 95
CANADIAN DOLLAR
Profit Taking Index
greater than 35.
ç BUY
Wave Four
channels holding
Oscillator pulled back
below zero. But stayed
above max allowed.
T-131
GET
Applying Technical Analysis
Expert Trend Locator – XTL
The need for XTL:
Advanced GET does an excellent job in identifying Elliott Wave counts. When combined with studies such as the Profit Taking Index, Wave Four Channels, Trend Channels etc., a very effective Mechanical trading strategy is available to GET users. This Mechanical strategy focuses mainly on:
a) Trading the Wave Four to Wave Five segment in Type One Trades.
b) Trading the end of Wave Five in Type Two Trades.
While this has been very productive, we still lacked a Mechanical type entry into some of the massive Wave Three swings generated in the market. So far, this was accomplished by staying in Type Two Trades entered at the end of a Wave Five and waiting for the Wave Three to develop. On numerous occasions, many of our users requested that we come up with a technique that allows early entries in Wave Three swings.
After some extensive research, we released the Joseph Trend Index in 1995. The JTI was designed to confirm Wave Three swings once the swing matured. It performed this task by producing an exponential curve in Major Wave Three swings. However, in some cases, the confirmation was accomplished a little too late for the user to take full advantage of the entire Wave Three swing, specifically in terms of an early entry.
The Task:
Since the last release of Advanced GET 6.0 in March 1995, I have dedicated my research exclusively to find a reasonable solution to identify Wave Three swings at its early stages. The result is XTL – Expert Trend Locator.
The Theory behind XTL:
If we carefully look at any chart, you will find periods where the market tends to trade back and forth without any clear direction with swings that reverse at random. Looking back, many of these swings can be classified as noise. The following example on the next page shows Feb 96
Gold trading sideways without any real Trend in the market.
For discussion purposes, this is a market that is trading in a random fashion. If you simply trade this market using a Trend following system, you would get whipsawed by buying at T-132
GET
Applying Technical Analysis
the highs and selling at the lows. Trend following methods generally do not work well in a sideways market. However, a trading approach using over bought/over sold indicators may work well in these periods.
Feb 96 Gold
Trading Sideways
However, as soon as you switch to an over bought/over sold trading strategy, the market will also switch to a Trending mode (or so it seems). This cycle never seems to change and can literally make a grown man cry. Once the market breaks into a Trend Mode, the prices trade in one direction with strong momentum and as the momentum continues, the swing is classified as a Major Trend. In Elliott Wave terms, this swing is labeled a Wave Three.
Feb 96 Gold
Trend Continues
Breakout î
T-133
GET
Applying Technical Analysis
The objective is to identify when the market is trading in a random fashion and when it is trading in a Major Trend. There are two additional requirements.
1) The Major Trend Mode has to be detected at any early stage so one can trade it 2) Minimize the false signals
We developed a Statistical model that checks for randomness in a given set of data. The model is somewhat similar to the Runs Test for Randomness (found in Advanced Statistical books).
Our Statistical model calculates a value between 0 and 1 for any given run of data. (0 being completely random and 1 being completely biased.) In addition to this, we also developed a way to measure a threshold level. If our test for Randomness for a given set of data exceeds this threshold level, we conclude that the market is potentially running or trading with a bias in the early stages of a Major Trend.
When a downward bias or Down Trend is detected, the bars are displayed in Red. When an upward bias or Up Trend is detected, the bars are displayed in Blue. If the Trend is neither Up or Down, the bars are displayed as usual in Black.
May 96 Crude Oil
with XTL (21 per)
Elliott Waves generated
by Advanced GET.
ì
BLUE BARS
shows Up Trend.
Regression Trend
Channels
î
ì
RED BARS
shows Down Trend.
T-134
GET
Applying Technical Analysis
Portability:
Any study is valid only if it is portable to other markets. The Statistical model we developed for the XTL meets this criteria. It works on all markets (Stocks and Futures) and for any time frame without any modification.
Compatibility with Elliott Wave analysis:
Almost all Wave Three swings generate strong and extended momentum to be classified as a Major Trend. In a majority of the cases, the XTL can identify the Major Trend at the early stages of a Wave Three. Since the XTL is a Statistical model and does not use Elliott Wave logic, it provides an independent confirmation and early entry tool for Wave Three type swings.
DESIGNATED USE FOR XTL:
Perhaps the single most valuable use for the XTL is to identify the early stages of a Wave Three type rally or decline. Elliott Wave analysis alerts the user of a potential Change in Trend and the XTL detects or identifies this Change in Trend. Thus the combination of Elliott Wave analysis and the XTL provides the added luxury of both anticipating and confirming the Trend.
On the next several pages, we will discuss how to use the XTL and explain some of the recommended trading rules. We suggest you use the training mode and test our recommended ideas on various markets. Then form a plan that fits your trading style.
SETTINGS FOR XTL:
There is only one user setting (input) required for calculating the Expert Trend Locator (XTL) study. This is the number of bars of data used to test for Randomness. The default setting is 35
periods and we highly recommend this. Using a 35 period setting reduces the number of false break out signals. However, using a 21 period setting allows the XTL to identify the Major Trend at a much earlier stage.
Thus, while the 35 period XTL offers less false signals, the 21 period XTL identifies the Major Trend one or two bars earlier.
Which setting (35 period or 21 period) you use is your personal preference. The best way to decide the best setting, use the training mode to back test on past data and find the setting that suits your style of trading.
We have also enclosed examples using both settings to further illustrate the advantages and drawbacks of both.
T-135
GET
Applying Technical Analysis
How to use XTL
Since XTL is a tool that identifies Major Trends, you can use it in a number of ways to enhance your trading. From our testing we have found the following method to be very reliable. When there is no Major Trend, the bars are plotted as regular bars in black. As soon as a Trend is detected, the color of the bar is changed to Red (Down Trend) or Blue (Up Trend).
BUY LEVEL
The first bar that changes the color
Add 1/2 of Bar length
is called the FIRST BREAK OUT
BAR.
ë
Add half the length of this Break
FIRST BREAKOUT
Out Bar to obtain a BUY LEVEL.
BAR
Buy (go long) when the market penetrates the BUY LEVEL provided the following holds true: a) XTL does not detect and display a bar with the opposite Trend color. In the above illustration, the buy was generated with an Up Trend signal.
Therefore, you should not see any Red color bars prior to the prices penetrating the BUY
LEVEL. Neutral (Black) bars are okay and does not alter the strategy. b) The STOP
is not taken out. (See next page for discussion of STOPS).
We suggest Adding 1/2 the length
of the Break Out Bar. Please feel
free to back test using the training
mode and find a length you are
BUY LEVEL
ç BUY
most comfortable with.
Add 1/2 of Bar length
For aggressive entries, you could
even set the BUY LEVEL just
above the high of the Break Out
ë
Bar.
FIRST BREAK OUT
BAR
You could enter the market with a
Buy Stop at the selected level.
T-136
GET
Applying Technical Analysis
In the following examples (unless stated otherwise) we will use a 35 period setting for calculating the XTL. We are also adding 1/2 the length of the Break Out Bar. (See Below)
This is the March 95, D-
Mark as of Feb 13, 1995.
March 95 D-Mark
The XTL has detected a
as of Feb 13, 1995
potential run in the market.
35 Per XTL
This bar is displayed in