Multicollinearity in the Trade

Multicollinearity - a statistical term for the problem, which is typical in technical analysis. The problem is that, unconsciously uses the same type of information many times. Technical analysts should be careful not to use technical indicators, which reflect the same type of information.

John Bollindzher, author of the famous Bollinger, said on that occasion: "The key to the successful use of technical analysis requires avoiding multicollinearity among the indicators. Multicollinearity - it is just repeated counting of the same information. Using four different indicators, whose values are obtained from the same of a series of closing prices to confirm each other is a clear example of such a mistake. "

The issue of multicollinearity is a serious problem in the technical analysis, where your money is in jeopardy. This is really a problem because collinear variables contribute redundant information and may make other variables seem less important than they actually do. One of the real problems is that multicollinearity is sometimes difficult to determine.

Technical indicators are to be divided into categories, to refrain from using too many of the same category. Below is a table in which the most common indicators are divided into categories:

The best way to quickly determine whether the indicator is collinear with another, is to display them graphically. Make sure you use a sufficiently large amount of data in the chart to get a high representativeness. If the indicators are rising and largely reduced in the same areas, it's likely that they are collinear, and you should only use one of them.

The first graph shows some examples of indicators that are collinear. Please note that all three indicators basically have the same information. If your analysis is that it was the supporting information, then you will fall into the trap of multicollinearity. Choose one of the indicators for your analysis, and do not use the other in the same analysis.

Below are some examples of indicators that are not collinear. These three are not similar at all and when they are interpreted correctly, each will give different information. It may confirm each other, but can not.


Conclusion
If you randomly select indicators to confirm his analysis, you are most likely, will fall into the trap of multicollinearity, using multiple indicators, which tells you the same information. They do not give you any further information. In fact, they limit your full understanding of the market situation. Do not look for confirmation of information to the collinear indicators, it will only introduce you to the confusion.