Moving averages versus MACD Moving averages have been a preferred tool for identifying and gauging trends since the early days of modern technical analysis, due to expectations that the prices will return to the mean. The introduction of the MACD over four decades ago provided a more strict set of rules for the otherwise loose set of averages traders employed. Both the averages and the MACD are excellent tools when it comes to trends, and traders use them when trends either exist are about to start.
Caution in Using Moving Averages and MACD Moving averages and, by extension, the MACD, are lagging indicators. This means that both of these tools are the last ones to reflect the change in the currency price. Calculating Moving Averages
Simple moving averages are calculated as the arithmetic mean of the closing prices and during the ongoing day; the “closing” price is actually the current price. The more days you use, the more insensitive the average will become to the current price action. If you want to allocate more weight to the current prices than to old closings, you can use a weighted moving average. And if you prefer to use more advanced moving averages, which are generally used in the MACD, then you will use exponentially-smoothed moving averages. Rather than looking at the formulas that you will never change, let’s look at these three averages on the same period so you can decide for yourself whether it’s worth using one over the other. In GFT’s DealBook® 360, you can apply exponential, modified, simple, triangular and weighted moving averages. To apply moving averages, access the new indicator menu from either the button on top of the chart window, or by right clicking anywhere on your chart. Select “Mov Avg Exponential,” Mov Avg Simple” and “Mov Avg Weighted,” and click OK to add these to your chart. Figure 1 shows the simple, weighted and exponentially-smoothed 21-day moving averages and Figure 2 displays the simple, weighted and exponentially-smoothed 55-day moving averages for AUD/JPY. You are looking at two Fibonacci-inspired averages for the (relatively) short and medium term. Although there are some differences, unless you are optimizing your analysis, you will probably be OK with any of these averages. 
Figure 1. The simple, weighted and exponentially-smoothed 21-day moving averages on AUD/JPY
Figure 2. The simple, weighted and exponentially-smoothed 55-day moving averages for AUD/JPY
Role of Moving Averages
Moving average alleviate market swings, as some medium-term traders prefer to limit the trading noise. Figure 3 shows the swings of the euro/yen uptrend being smoothed by a 21-day moving average. 
Figure 3. A 21-day moving average smoothes the swings of the euro/yen uptrend
Moving averages should be used as flexible support and resistance lines. To this end, they should use the intersection between the currency and a short moving average, anywhere between 10 and 30 days, to obtain buy and sell signals in trending markets. For instance, the AUD/JPY chart in Figure 4 shows the intersections between the cross and the 21-day moving average; there is a selling signal (red arrow), a buying signal (blue arrow) and support provided by the average (green arrow). You may choose to go long on the currency pair when it crosses above such a moving average and hold this position in a trending market for as long as it holds above this average. Conversely, you could go short after a currency pair falls below the support of a moving average. A close above or below the average will confirm the crossover, but savvy traders may like to enter (or exit) these positions as soon as the crossover takes place. 
Figure 4. The intersections between the cross and the 21-day moving average provide a selling signal (red arrow) and a buying signal (blue arrow). The green arrow identifies the area of support provided by this average.
Another benefit of using moving averages is the overbought and oversold signals. If the currency is moving too far away from the moving average, this is a warning that the currency might change direction. However, a divergence is only a condition, not a signal, so you must employ a trigger from a different source, such as the break below a support or a trend line before you trade. Figure 5 shows the sterling/dollar being initially oversold (see the blue arrow) and then overbought (see the two red arrows) versus the 21-day moving average. In all three instances, the sterling/dollar approached the average, providing you with three opportunities to trade. Be forewarned that there is a less exciting way for a currency pair to alleviate its overbought or oversold condition: trade sideways while the lagging moving average catches up. 
Figure 5. Sterling/dollar is initially oversold (see the blue arrow) and then is overbought (see the two red arrows) versus the 21-day moving average.
The divergences described so far between a currency and the moving average provides short-term divergence signals. To identify medium-term overbought or oversold conditions, traders should use either weekly charts or compare two different moving averages. Try GFT's free forex trading software, now. Figure 6 shows four instances where the sterling/dollar was overbought on a weekly basis (see red arrows). 
Figure 6. There are four instances where the sterling/dollar was overbought on a weekly basis (see red arrows) Figure 7 shows the daily euro/Swiss franc being overbought in the medium-term five times (the red arrows), as the 21-day moving average was too far away from the 55-day moving average. The cross promptly approached or broke the 21-day moving average and alleviated its medium-term overbought condition. Based on this method, at the end of this chart the cross is likely to decline. 
Figure 7. The daily euro/Swiss franc was overbought in the medium-term five times (see the red arrows), as the 21-day moving average was too far away from the 55-day moving average
A common reason to use averages is to view the intersection between two moving averages, which is called the double-crossover method. Given that averages are lagging tools, these crossovers tend to act better as confirmation signals rather than entry or exit signals. On a combination of two moving averages, a buying signal occurs when the shorter of two averages intersects the longer one upward. This strategy provides many false breakouts. As Figure 8 shows, the red arrow points to a sell and the blue arrow points to a buy signal, as defined by the cross of the two-day moving average above the 55-day moving average. The first signal is simply wrong, while the second signal is seriously late. 
Figure 8. The red arrow suggests a sell of the euro/Swiss franc cross and the blue arrow points to a buy signal, as defined by the cross of the two-day moving average above the 55-day moving average. The first signal is wrong, while the second signal is very late.
To filter out the crossover between moving averages, you should use a variation introduced by the Japanese technicians: a dead cross and a golden cross. A dead cross is formed by the intersection of two consecutive moving averages that move in opposite directions. It has a high rate of failure, so it’s fair to assume that is technically insignificant. In Figure 8, the dead cross is identified by the red arrow. The intersection of two moving averages that move in the same direction is called a golden cross. This intersection provides a more reliable signal that the currency will continue moving in the same direction. You can see an example of a golden cross in the euro/Swiss franc in Figure 8 as identified by the blue arrow. The triple crossover method involves three moving averages. The intersection of short and middle averaged provides a warning signal, and the intersection of the middle and long averages give a trading signal. Let’s take the example of the signals generated by the combination of the fast 5-, 20- and 60-day moving averages. The warning occurs when the 5-day moving average crosses the upward moving 20-day average. The buying signal occurs when the 20-day moving average crosses the 60-day average upward. Number of Moving Averages Generally, traders should use at least one and as many as four moving averages in order to capture signals in different time horizons. There are no general suggestions in terms of the number of averages to use or their length. Some of the more popular combinations are 5-9-18 days for the very short-term oriented traders and 10-20-60 for the medium-term traders. In addition to plotting them on daily charts, traders use moving averages other periods, whether short-term or long-term: 15-minute, hourly, weekly, etc. Moving Average Convergence Divergence indicator (MACD)
Gerald Appel developed the moving averages convergence divergence (MACD) to measure the trend momentum and to point to entry and exit points and gauge the market with signals generated by diverging consecutive. This oscillator uses exponentially smoothed moving averages. The MACD oscillator can be added within GFT’s DealBook® 360 by accessing the new indicator menu from either the button on top of the chart window, or by right-clicking anywhere on your chart. Select MACD from the list, and edit any parameters. The MACD can also be applied within DealBook® WEB. After opening a new charting window, select MACD from the drop-down indicators window to apply the oscillator in a subgraph. The MACD oscillator is formed of two lines: the difference between two exponential moving averages on 12-day and 26-day, and a 9-day exponential moving average which acts as a trigger. This oscillator moves on an open scale around the zero line. The MACD can be represented on the chart as a line or as a histogram; by default it is displayed in DealBook® 360 by both a line and a histogram in a subgraph. The MACD provides buying signals when: 1. It rises above the zero line
2. The trigger line is above the line that marks the difference between the 12-day and 26-day moving averages
3. There is a bullish divergence with the currency In reverse, the MACD shows selling signals when:
1. It falls below the zero line
2. The trigger line crosses below the difference between the 12-day and 26-day moving averages
3. There is a bearish divergence with the currency As Figure 9 shows, the MACD provided two buying signals (blue arrows) and two selling signals (see the red arrows) on the euro/dollar daily chart. 
Figure 9. MACD applied to euro/dollar. The oscillator provided two buying signals (blue arrows) and two selling signals (see the red arrows).
Duration of the Averages for MACD
The MACD uses the following numbers: 9, 12 and 26. Where do these numbers come from? When this oscillator was designed back in the 1960s, the trading week was six days long. These parameters therefore represent one and a half business weeks, two business weeks, and one business month respectively. Since the trading week is only about five days long, one may want to modify the MACD parameters to 7 from 9, to 10 from 12, and to 22 from 26. In this case, the chart from Figure 9 would be slightly different in Figure 10 – the trading signals will be seen faster. 
Figure 10. The initial MACD parameters reflected the length of the old work trading and business month. Since the trading week is about five days, the MACD parameters should be adjusted to 7 from 9, to 10 from 12, and to 22 from 26. The trading signals will be seen faster.
Figure 11 shows a combination of the 21- and 55-day moving averages and the adjusted MACD on the dollar/yen chart. The crossovers between the two averages provide two Golden crosses (one down and one up), while the MACD provides three selling signals (see the red arrows) and two buying opportunities (marked by red arrows). 
Figure 11. A combination of the 21- and 55-day moving averages provides two Golden crosses (one down and one up and the adjusted MACD provides three selling signals (see the red arrows) and two buying opportunities (marked by red arrows) on the dollar/yen chart.
A combination of moving averages and the MACD will be helpful in gauging the momentum of the currency pair and in identifying point overbought and oversold conditions. However, traders must use these tools in addition to in-depth trend analysis and pattern recognition.
| 
