CHARTING THE MARKET
Moving Averages
If trendlines are the first thing that most traders and technical analysts learn to do with a chart, moving averages are perhaps a close second. Rereading swing trader Dave Landry's commentary about a moving average-based trading system he developed back in the mid-1990s, I saw one concise explanation of why -- in a world of ever more complex approaches to trading the markets -- moving averages continue to matter:
As technicians, we are all guilty sooner or later of trying to reinvent the wheel with elaborate technical analysis. We'd all like to find the Holy Grail of an indicator that will make us rich. I have spent countless hours searching for such an indicator, only to rediscover something far less complex. The truth is this: There is no perfect indicator or system that will lead to riches while avoiding all risk.
While Landry is quick to add he is not condemning the more complex or esoteric technical approaches to analyzing and trading the markets, he does underscore the fact that even the most elaborate and discriminating methods (some of which are presented below) are all derived from the more basic technical approaches. As far as basic technical approaches go, moving averages are almost as basic as you can get.
Writing about moving averages in his classic Technical Analysis Of The Futures Markets, John Murphy observes that "the moving average is one of the most versatile and widely used of all technical indicators. Because of the way it is constructed and the fact that it can be so easily quantified and tested, it is the basis for most mechanical trend-following systems in use today." Per Murphy, moving averages are among the best weapons that the trader (or investor) has in combating the inherent subjectivity in analyzing price action. Because they are trend-following in nature, moving averages help traders accomplish the first -- most important -- feat in trading: getting and staying on the right side of the trend.
How do moving averages work? A moving average is exactly that -- a running average of a sum of N numbers. For instance, a 10-day moving average of price closes represents an average of the past 10 days' values at their closes. The average is referred to as a "moving" average because each day, a new day's value at the close is added and the day that is now 11 days ago has its values dropped from the calculation. Thus, each day in a moving average (or, more accurately, each bar, insofar as moving averages can be taken of hours, days, weeks, or even discrete five-minute periods) has a new day's value being added to the calculation and an old day's value being dropped.
As Landry suggests, there are plenty of wrinkles that technical analysts have added in an attempt to get more out of the classic moving average. One of the most common twists is the exponential moving average (EMA). The difference between the simple moving average (SMA) and the EMA is that the exponential version tends to respond quicker to newer prices, thus eliminating some measure of lag that critics of the simple moving average tend to object to (Figure 1). While the actual computation of the exponential moving average is too involved to be briefly described here, understand that by weighing more recent data heavier than older data, the EMA has brought even more traders to the moving average camp.
Figure 1: Breaking out at 1.1690, EUR/USD hugs its 20-bar exponential moving average as it advances to 1.1715.
There are two other types of moving average to mention -- and this is without even discussing the use of combined moving averages of different lengths. These two types are the linearly weighted moving average and the triangular moving average. The linear weighted moving average attempts to do what the EMA does in the way of emphasizing recent data relative to older data. The linear weighted moving average multiplies each day (or bar) of the moving average by its ordinal value (that is, the fifth day is multiplied by 5, the fourth day is multiplied by 4, and so forth. This total is then divided by the sum of the multipliers -- the multipliers in a 10-bar moving average would add up to 55, the multipliers in a 20-bar moving average would add up to 210, and so forth).
The triangular moving average is another, more complex spin on the moving average. The triangular moving average works by double-smoothing price action and is, in essence, a moving average of a moving average. According to Steven Achelis, author of Technical Analysis From A To Z, a triangular moving average actually puts most emphasis on the middle portion of data in the moving average calculation, as opposed to the most recent data as with the exponential moving average.
--David Penn, Technical Writer
Learn more about moving averages!
Achelis, Steven B. [2000]. Technical Analysis From A To Z, McGraw-Hill.
Arrington, George R. [1991]. "Building A Variable-Length Moving
Average," Technical Analysis of STOCKS & COMMODITIES, Volume
9: June.
_____ [1993]. "The JSA Moving Average," Technical Analysis
of STOCKS & COMMODITIES, Volume 11: October.
Lafferty, Patrick [1995]. "The End Point Moving Average," Technical
Analysis of STOCKS & COMMODITIES, Volume 13: October.
Landry, David [1996]. "The 2/20-Day EMA Breakout System," Technical
Analysis of STOCKS & COMMODITIES, Volume 14: December.
McGuinness, Charles [1990]. "Optimizing Moving Averages," Technical
Analysis of STOCKS & COMMODITIES, Volume 8: April.
Murphy, John J. [1986]. Technical Analysis Of The Futures Markets,
New York Institute of Finance.
Pring, Martin J. [2000]. "Using The Simple Moving Average," Technical
Analysis of STOCKS & COMMODITIES, Volume 18: June.
Roth, Raymond [1992]. "Understanding Exponential Moving Averages,"
Technical Analysis of STOCKS & COMMODITIES, Volume 10: August.
Schmidt, Heidi [1988]. "Moving Averages Made Simple," Technical
Analysis of STOCKS & COMMODITIES, Volume 6: March.
Wayman, Robert [1995]. "Whipsaws, Linear Trends & Simple Moving
Averages," Technical Analysis of STOCKS & COMMODITIES,
Volume 13: October.
Originally published in the December 2003 issue of Technical Analysis of STOCKS & COMMODITIES magazine. All rights reserved. © Copyright 2003, Technical Analysis, Inc.
Return to December 2003 Contents