Technical AnalysisTrading Systems

Building A Trading Indicator

Elliott and Gann have become household names among the worldwide trading community. These pioneers of technical analysis developed some of the most widely used techniques in the field. But how did Ralph Nelson Elliott and W.D. Gann come up with these techniques, and how did they become so successful? Truth be told, it’s not as difficult as it sounds! This article takes you through the process of building your own custom indicator, which you can use to gain an edge over the competition.

Background
Recall that the theory behind technical analysis states that financial charts take all things into account – that is, all fundamental and environmental factors. The theory goes on to state that these charts display elements of psychology that can be interpreted via technical indicators.

To better understand this, let’s look at an example. Fibonacci retracements are derived from a mathematical sequence: 1, 1, 2, 3, 5, 8, 13 and so on. We can see that the current number is the sum of the previous two numbers. What does this have to do with the markets? Well, it turns out that these retracement levels (33%, 50%, 66%) influence traders’ decisions to such an extent that the levels have become a set of psychological support and resistance levels. The idea is that, by finding these points on charts, one can predict the future directions of price movements.

Components of an Indicator
All indicators are created to predict where a price is likely to head when a certain condition is present. Traders try to predict two basic things:

  • Support and resistance levels: These are important because they are the areas at which prices reverse direction.
  • Time: This is important because you need to be able to predict when price movements will occur.

Occasionally, indicators predict these two factors directly – as is the case with Bollinger Bands or Elliott’s Waves – but indicators commonly have a set of rules enacted in order to issue a prediction.

For example, when using the breadth thrust indicator (which is represented by a line indicating momentum levels), we need to know which levels are relevant. The indicator itself is simply a line. The breadth thrust indicator looks similar to RSI, in that it is “range-bound,” and it is used to gauge the momentum of price movements. When the line is in the median zone, there is little momentum. When it rises into the upper zone, we know that there is increased momentum and vice versa. One could look to take a long position when the momentum is on the rise from low levels and look to short after the momentum peaks at a high level. It is important to set rules to interpret the meaning of an indicator’s movements in order to make them useful.

With this in mind, let’s look at ways of creating predictions. There are two main types of indicators: “unique” indicators and “hybrid” indicators. Unique indicators can be developed only with core elements of chart analysis, while hybrid indicators can use a combination of core elements and existing indicators.

Components of Unique Indicators
Unique indicators are based on inherent aspects of charts and mathematical functions. Here are two of the most common components:

  1. Patterns
    Patterns are simply repeating price sequences apparent over the course of a given time period. Many indicators use patterns to represent probable future price movements. For example, Elliott Wave theory is based on the premise that all prices move in a certain pattern. A simplified example is shown in Fig 1:

Fig 1 – An Elliot Wave Pattern

There are many other simple patterns that traders use to identify areas of price movement within cycles. Some of these include triangles, wedges, and rectangles.

These types of patterns can be identified within charts simply by looking at them; however, computers offer a much faster way to accomplish this task. Computer applications and services provide the ability to automatically locate such patterns.

  1. Mathematical Functions
    Mathematical functions can range from price averaging to more complex functions based on volume and other measures. For example, Bollinger Bands are simply fixed percentages above and below a moving average. This mathematical function gives a clear price channel showing support and resistance levels.

Components of Hybrid Indicators
Hybrid indicators use a combination of existing indicators and can be thought of as “simplistic trading systems”. There are countless ways in which elements can be combined to form valid indicators. Here’s an example of the MA crossover:

This hybrid indicator utilizes several different indicators including three instances of the moving averages. One must first draw the 3-day, 7-day and 20-day moving averages based on the price history. The rule then looks for a crossover in order to buy the security or a cross-under in order to sell. This system indicates a level at which price movement can be expected and provides a reasonable way to estimate when this will occur (as the lines draw closer together). Here’s what it might look like:


Fig 2 – A Moving Average Crossover – Source: Yahoo Finance

Creating an Indicator
A trader can create an indicator by following several simple steps:

  1. Determine the type of indicator you wish to build: unique or hybrid.
  2. Determine the components to be included in your indicator.
  3. Create a set of rules (if necessary) to govern when and where price movements should be expected to occur.
  4. Test your indicator in the real market through backtesting or paper trading.
  5. If it produces good returns, put it into use.

An Example
Suppose we want to create an indicator that measures one of the most basic elements of the markets: price swings. The goal of our indicator is to predict future price movements based on this swing pattern.

Step 1:
We look to develop a unique indicator using two core elements, a pattern and math functions.

Step 2:
Looking at weekly charts of company XYZ’s stock, we notice some basic swings between bullishness and bearishness that each last about five days. As our indicator is to measure price swings, we should be interested in patterns to define the swing and a mathematical function, price averages, to define the scope of these swings.

Step 3:
Now we need to define the rules that govern these elements. The patterns are the easiest to define: they are simply bullish and bearish patterns that alternate every five or so days. To create an average, we take a sample of the duration of upward trends and a sample of the duration of downward trends. Our end result should be an expected time period for these moves to occur. To define the scope of the swings, we use a relative high and a relative low, and we set these at the high and low of the weekly chart. Next, to create a projection of the current incline/decline based on past inclines/declines, we simply average the total inclines/declines and predict the same measured moves (+/-) occur in the future. The direction and duration of the move, again, is determined by the pattern.

Step 4:
We take this strategy and test it manually, or use software to plot it and create signals. We find that it can successfully return 5% per swing (every five days).

Step 5:
Finally, we go live with this concept and trade with real money.

In Summary
Building your own indicator involves taking a deeper look into technical analysis and then developing these basic components into something unique. Ultimately, the aim is to gain an edge over other traders. Looking at Ralph Nelson Elliott or W. D. Gann, their successful indicators gave them not only a trading edge but also popularity and notoriety within financial circles worldwide.

Justin Kuepper can be contacted at Internationalinvest

Justin Kuepper has over 15 years of experience in the market as a private investor and financial journalist.

He spent several years building and managing financial portals before obtaining a position with Accelerized New Media, owner of SECFilings.com, ExecutiveDisclosure.com and other popular financial portals.

Justin still writes on both finance and technology topics.

Justin Kuepper has over 15 years of experience in the market as a private investor and financial journalist.He spent several years building and managi...

0007

Senior member
2,357 644
It's an interesting article with an idea rarely suggested. Many articles about indicators seem to spend most of their time explaining why this indicator is better than another and then go into meticulous detail of how easy it is to become rich by just slavishly following that particular indicator. Here's a guy who tells you to understand what you're looking for and then make yourself an indicator to help find it. It's something I've been doing for quite a while now and it is very rewarding.

Having spent much time exploring the use of every indicator known to man as well going through the time tested process of fitting them all on one chart simultaneously I came to the conclusion that something better was needed. So I tried a kind of reverse-engineering of the indicator usage process. Many people have remarked that you need lots of screen time to become proficient in assessing a trade and after much practice I found that to be so. I found that a certain kind of chart pattern/setup of price and very basic lines e.g. moving averages (and most importantly their parameters such as slope / difference / direction) could indicate a very high probability of a decent trade. What I then had to do was construct a hybrid scan/indicator which alerted me to such situations. So for instance, using the SP500 constituents a mountain of 500 possible trades could easily be reduced to a couple of dozen on a good day and much less on others. Without having to plough through 500 charts and suffering the equivalent of "Death by PowerPoint" it becomes much easier to sort out your good trades. With the current volatility of the market there are lots of opportunities & this procedure is paying off well .

If you decide to take this approach you will need a charting software that allows scripting/programming and that in itself will provide you with a learning task which will probably involve a load of hard work but in this game if you've got that far you shouldn't be surprised at that. I use an ancient copy of Metastock (version 9) which I bought years ago and has been great. If I were starting today I probably also look at Amibroker which is excellent value and can be made to do almost anything If you are clever enough. You could also use very much more expensive solutions which would probably be just as good!

Once you've got your trade identified (using free EOD or delayed pricing – quality real-time data can be expensive) you can monitor it on your broker's charting in real time if you want to – a worthwhile procedure at the moment with opening gaps et cetera. I use IG 's free charting which is excellent and just look in 3 times a day – at the open, midway, & the close: after all, you don't want to become a screen slave.

If you're prepared to put in the work then writing your own indicators/scans can be very worthwhile. Incidentally, if you're inclined to this way of thinking then it's worth reading Andreas Clenow's books.
 
Last edited:
  • Like
Reactions: cantagril

NVP

Legendary member
37,283 1,936
Hey 0007

Great post....and some great advice here....

People hopefully will pick up on the slope strength comment...as it's so powerful an area for identifying high prob trades when combined with other triggers......

That particular element has been in my own mix of trading forex for nearly 20 years......and is powerful stuff...

N