The Average Directional Index indicator (ADX)

To continue the article series about the objective technical analysis, I will discuss the Average directional index (or ADX) in this article. Earlier, I have written articles about the moving average, the moving average convergence/divergence indicator and the relative strength index, which all are objective technical analysis methods. Just like the other objective technical indicator, the average directional indicator is used to measure trends of a share.

The ADX indicator is build up out of three different components. These components are the true range, minus directional indicator and the plus directional indicator. These components are used to calculate the ADX. But before I explain the calculation of the ADX, I will explain the different components.

First, the true range (TR). The TR is defined as the largest outcome of the three following options:

• Difference between period (usualy day) high and the period low
• Difference between period high and previous period close
• Difference between period low and previous close

Because the TR should be positive, the smallest value must be deducted from the highest value for each of the previous options.

After having calculated the TR, the average true range (ATR) can be calculated. The ATR is an indicator in itself which measures the volatility of a share. The ATR is the average of the (usually) 14 latest TR values. As said, you can use the ATR as an individual indicator, but I will come back to that in a later article.

Second, the minus directional indicator (-DI). This indicator is measured by using the minus directional movement (-DM) of the previous and current period lows and the previous and current period highs. The -DM is the previous period low minus the current period low, but only when this difference is greater than the previous period high minus the current period high. This makes the following calculation for the -DM:

-DM = previous period low - current period low

The plus direcional indicator (+DI), is calculated exactly the same. So the +DI is based on the plus directional movement (+DM) for which the lows and highs are used the other way around. So the calculation is:

+DM = previous period high - current period high

In addition to the calculation of the -DM and +DM is, that if the result of the calculation is less than zero, the indicator is zero. Besides that, only the largest directional movement difference counts. So when the -DM is larger that the +DM result, the +DM will be put to zero. And of course the other way around.

To smooth the outcomes of the directional indicators, a 14 period timeframe is being used. So the previous 14 periods are added, which creates both an higher +DM and -DM. The calculation of both the -DI and +DI is done using the following formulas:

-DM14 = previous 14 periods -DM

+DM14 = previous 14 periods +DM

-DI14 = 100 x (-DM14 / ATR)

+DI14 = 100 x (+DM14 / ATR)

Now that you know how to calculate all components of the ADX, I can explain the calculation of the ADX. Because ADX means Average directional index, you could guess that we first have to calculate the directional index (DX). To calculate the DX, the sum and differences of both the directional indicators are needed. The difference should always have a positive result, so you have to deduct the smallest directional indicator from the largest. So we need to perform the following calculations:

DI14 difference = -DI14 - +DI14 or DI14 difference = +DI14 - -DI14

and

DI14 sum = -DI14 + +DI14

To calculate the DX, we must perform the following calculation:

DX = 100 x (DI14 difference / DI14 sum)

Now to calculate the ADX, we have to take the average DX from the last 14 periods. The calculation will be:

ADX = previous 14 period DX / 14

In short: the ADX measures the strength of a trend, not necessarily the direction.

When the ADX is above 25, the share is assumed to be trending strongly. When the ADX is below 20, the assumption is that the share does not show a trend.

Because you can both put the -DI and +DI indicator in the same graph as the ADX, you can also see the crossovers of the -DI and +DI. So, besides the actual ADX value, you can get another indication out of the graph. When the -DI crosses the +DI, a trendturning signal is given. You can also find the movement of the trend. When the -DI is above the +DI, the trend is downwards. When the +DI is above the -DI, the trend is upwards.

What are the pitfalls of the ADX?

Because the ADX is calculated by using different averages which take some period to calculate, there is some timelag in the result of the ADX. The longer periods are taken, the more lag there is in the ADX. This is the main pitfall of the ADX.

Besides the pitfall mentioned, I will add (again) that you never should trust only one technical indicator. You should always rely on multiple signals which correspond with each other. This will reduce the number of false signals.