Using a moving average on a price chart of an index or exchangetraded fund (ETF) indicates to an investor which side of the market to be on. This approach will enable an investor to capture approximately 75 percent of the move in each direction. Expect to miss the first part of a trend change because the moving-average line lags the market, being based on historical data. When a new movingaverage signal is given, the odds are in your favor that you will make money if you invest with the new trend.
SIMPLE MOVING-AVERAGE BASICS
A simple moving average is calculated each day by adding the last day’s closing price to the total of all closing prices over the time period chosen and then subtracting the first day’s closing price. However, you need not calculate it because all the Internet charting Web sites automatically do it for you once yo
A simple moving average is calculated each day by adding the last day’s closing price to the total of all closing prices over the time period chosen and then subtracting the first day’s closing price. However, you need not calculate it because all the Internet charting Web sites automatically do it for you once you indicate the time frame you wish to graph.
Securities prices, market indexes, and mutual-fund prices vacillate up and down from day to day, week to week, and month to month. It is often difficult to discern which way the prices are actually moving. The moving average is used to smooth the data so that the trend can be easily detected. A rising moving-average line indicates that prices are trending up, whereas a declining line indicates the opposite. A flat line indicates a market stuck in a trading range that can’t seem to make up its mind where it is going. Frequently, the market is in a fixed trading range bouncing between the upper and lower limits (price levels) for an extended period of time. The smart investor waits for this trading range to be pierced either up or down before making a move in the direction of the breakout.
You can create a moving average of any length (e.g., 9-, 25-, 65-, or 100-day moving average) and for any time period (minutes, hours, days, weeks, or months) depending on what you are trying to achieve, for example, for trading for day trading (minute charts) or for intermediate investing (daily and weekly charts) and for long-term investing (weekly and monthly charts).
Basically, when timing the market, you want to be long (invested) the market at the instant when the price of the stock or index crosses above the moving-average line and out-of-the-market (in cash) or short the market when the price of the stock or index crosses below its moving-average line. The major market indexes, such as the Dow Jones Industrial Average (DJIA), the Nasdaq Composite, the Standard & Poor’s (S&P) 500 Index, or the Wilshire 500 Index, can be used to determine the market trend using the moving-average approach.
50-Day Moving Average (dma)
Figure 9-1 (which shows both the 50- and 200-dmas for comparative purposes) shows that when S&P 500 Index’s price crossed below the 50-day moving average (line closest to index price) in June and September 2008, you would have exited the market, missing the bulk of the devastation from September 2008 through March 2009. And at the end of March 2009, the index crossed the 50-dma to the upside, indicating a buy signal on the market. However, as you can see, there were whipsaws above and below this moving-average line in June, July, and November 2009, followed by more whipsaws in 2010. This would have resulted in numerous buy and sell signals, resulting in commission costs and some small losses because the
trend was up during this period. Whipsaws are a common feature of moving averages, but such is the price you pay to stay on the right side of the market. Once the clear trend is established, it will continue until a trend reversal occurs.
200-Day Moving Average
The 200-dma is most indicative of the long-term trend because it covers 40 weeks. Viewing the same chart in Figure 9-1 but now looking at the 200-dma, you can see that the index was below this moving average at the beginning of 2008. The index actually crossed it in December 2007, but it was blocked by the description in the left corner of the chart. Thus you would have had a market sell signal in December 2007, missing the entire brutal 2008 bear market. The first buy signal occurred in early June 2009, then two minor whipsaws occurred, then another buy signal occurred in early July, and then other whipsaws occurred in May through September. Thus even the longer 200-dma has whipsaws as the market rests before making its next move.
Unfortunately, no one can predict whether that move will be up or down, so just pay attention to what the charts are saying, and follow the indicator if you are using it in your arsenal of tools. Some of the more popular moving-average periods used by market professionals are the 20-dma, the 50-dma, and the 200-dma.
Using the Single Moving Average as an Investing Trigger
If you are using the S&P 500 Index with a single moving average, then you could invest in an ETF of your choice (such as SPY or DIA) when the index price rises above the moving-average line (e.g., 50- or 200-dma). Likewise, you would sell the ETF when the index price declines below its moving-average line. When you sell, you can either place the proceeds into a money-market fund or, if you are an aggressive investor, go short using a no-load bear (inverse) mutual fund offered by Rydex or ProFunds or using the inverse ETFs from RydexShares™ or ProShares. This is the way you would handle any investment vehicle, whether it is a stock, an index fund, a sector fund, or an ETF. Whenever the price crosses above or below the moving-average line, you act on that signal.
Dual Moving Averages
One interesting variation of the moving-average approach is to use a combination of two different moving-average time periods. This is often referred to as a dual-moving-average crossover system. One of the most well-known and widely followed combinations is the 50/200-dma. In this case, a buy signal is given when the 50-dma crosses over the 200-dma (known as the golden cross), and a sell signal is given when the 50-dma falls below the 200-dma (known as the death cross). The difference between the single-moving-average system and the dual-moving-average system is that in the single system the trading signals are given when the price crosses the moving-average line, whereas in the dual system the signal is given when the faster moving-average line (50-dma) crosses the slower moving-average line (200-dma). Also, dual-moving-average signals tend to be slower than single-moving-average signals.
In Figure 9-1 both the golden cross and the death cross are illustrated, which is quite rare in such a short time period. Note the golden cross buy signal in late June 2009 and the death cross sell signal in early July 2010 and finally the golden cross buy signal again in late October, all the way on the right side of the chart. The research on the performance of the stock market three months, six months, and twelve months after each type of cross is mixed. During the July to October 2010 period, there were numerous articles appearing in the press and on the Internet written about the significance of these crossovers, but most of the well-known researchers are not that impressed with the track record of crossovers in forecasting future stock market performance.
RESEARCHERS’ MOVING-AVERAGE TESTS
Over the years, a number of researchers have tested various moving- average lengths. This section focuses on the work of Robert W. Colby, Michael McDonald, Paul Merriman, John J. Murphy, and Mark Courter.
Robert W. Colby’s Moving-Average DJIA Tests
Robert W. Colby is the author of The Encyclopedia of Technical Market Indicators (McGraw-Hill, 2003). This 820-page book covers hundreds of technical market indicators that Colby backtested against a buyand- hold strategy. As you may recall from previous chapters, backtesting involves using historical data to determine how an index of stocks would have performed in the past using a particular strategy. The results do not mean that anyone could have achieved the results shown. Basically, backtesting is used to test specific strategies to determine their validity on a theoretical basis. If the results are valid over long time frames, then they may work as well in the future.
Colby tested many simple moving-average lengths using DJIA daily closing prices from 1900 through 2001. His assumptions in testing the data were that all profits were reinvested but that dividends, transaction costs, and taxes were excluded from the calculations. A buy signal was generated when the DJIA price pierced the moving average (being tested) to the upside, and a sell signal was generated when the DJIA price declined below the moving average to the downside. All sell signals were used to short the market. Surprisingly, Colby found that all the simple moving averages between 1 and 385 days beat buy-and-hold. In particular, he discovered that all the moving averages between 1 and 100 days beat buy-and-hold by more than a ratio of 7:1.
Colby’s best results were with the shortest lengths, the 1-, 4-, and 5-dmas, which produced outstanding results with net profits in the billions of dollars. For the intermediate-term lengths, the 66- dma was 31 times better than buy-and-hold over the 102-year test period. The price crossovers of the 66-day simple moving average generated a net profit of $639,933 on a $100 initial one-time investment compared with $20,105 profit for buy-and-hold. Testing the popular 200-dma, Colby determined that the net profit was only $121,257, about 19 percent of the 66-dma’s profit but still six times better than buy-and-hold.
Of the long-term period lengths, Colby determined that the 126-dma was the most profitable strategy, with a net profit of $426,746. This was 2,022.54 percent more profitable than buy-and-hold. Colby also noted that since the crash of 1987, the short-selling strategy on the sell signals has not been profitable. Interestingly, only 185 of 886 trades (21 percent) were profitable over this 102-year time frame. But the average winning trade amount of $6,886.30 compares very well with the average trade loss of approximately $1,238.08. This is why, overall, this strategy is profitable. The average number of days for each trade’s length is 42.
You may be wondering why Colby’s test results for the shortterm moving averages had much higher returns than those for the intermediate- and longer-term moving averages. Clearly, the faster moving averages (fewer days) react more quickly to changes in direction than longer-term moving averages; therefore, overall, the profits keep mounting and the losses are cut short more quickly. One downside of using short-term moving averages is the large number of trades during the year. For the average investor, an intermediate- to long-term moving-average approach is much easier to implement and track and much less expensive in commission brokerage charges.
Michael McDonald’s Moving-Average S&P 500 Index Tests
Michael McDonald, in his book, Predict Market Swings with Technical Analysis (Wiley, 2002), backtested numerous moving-average lengths to determine which ones provided the best performance compared with buy-and-hold on the S&P 500 Index, including reinvested dividends. He found that superior results were obtained using the 72- and 132-dmas. Remember that Colby, who did his backtesting with the DJIA, found that the 66- and 126-dmas were high-profitability strategies. Results this similar on two different indexes for different time periods provide confirmation that these moving-average lengths are valid timing tools.
In 1983, McDonald backtested all daily moving averages using the S&P 500 Index between 5 and 200 days in 1-day increments to determine the most profitable time frame. McDonald used the next day’s closing price, after the signal was given, as the buy or sell price. This approach is a more conservative one than using the same-day price when the moving average crossovers occurred because acting on the same-day price usually provided increased profits for the investor. McDonald replicated his research in 1992 and in 1999 to confirm the results of his original study. To do this, he divided the data into four distinct periods: the Great Depression, the 20-year post–World War II bull market, the 1966–1982 sideways market, and the 1982–1999 bull market. According to McDonald, the original results were confirmed by the later studies.
Over the 71-year total test period, the buy-and-hold strategy produced an average return of 10.3 percent a year, including reinvested dividends, whereas the shorter moving averages of 50 days or less produced average returns of 9.5 percent a year, not keeping up with buy-and-hold. One of the best returns was obtained with the 130-dma, which provided an annual return of 12.5 percent. Remember that Colby’s short-term moving-average results did not include reinvested dividends and used the DJIA instead of the S&P 500 Index; therefore, his results were different.
Paul Merriman’s 100-Day Moving-Average Nasdaq 100 Index Tests
Paul Merriman, president of Merriman, Inc., is a well-known investment advisor and educator who has conducted extensive backtesting on market timing and diversification strategies. Merriman wrote an article on April 20, 2001, entitled, “All About Market Timing,” that focused on the 100-dma to make buy and sell decisions. You can read it on his Web site at www.fundadvice.com/ fehtml/mtstrategies/0104.html.
Merriman tested what would have happened if you had invested $1,000 at the beginning of 1942 and left the money untouched through year end 2001 compared with using a 100-day simple moving average to move in and out of the market. The simple 100-dma system was used to generate buy and sell signals on the S&P 500 Total Return Index (dividends reinvested) and on the Nasdaq 100 Index. The signals were initiated when the index prices crossed their respective moving averages to the upside (a buy signal) and the downside (a sell signal).
Merriman compared the results with a buy-and-hold strategy and a 50 percent (equivalent to a beta of 1.5) and 100 percent leveraged position (a beta of 2.0). On all sell signals, the proceeds were invested into a money-market account. Merriman did not go short on the sell signals. The data in these original studies ended at year end 2000, but Merriman was kind enough to provide me with the updated data through year end 2001 (more up-to-date data through 2009 will be available on the Web site by the time this book is published). The results of the Nasdaq Index testing are presented here.
Merriman tested the 100-dma using the Nasdaq 100 Index over a 30-year period from 1972 to 2001 with and without leverage. From 1972 through 1999, this timing approach had only one loss in 1993 of –1.7 percent, this compared with five years of losses (1973, 1974, 1981, 1984, and 1990) for buy-and-hold. Moreover, even though 2000 was a big down year for buy-and-hold at –36.8 percent, the moving-average timing lost merely –19 percent. In 2000, buy-and-hold lost –32.7 percent, whereas the 100-dma was up 1.7 percent. This is quite an accomplishment.
The buy-and-hold strategy for the Nasdaq 100 Index from 1972 through 2001 yielded an annualized return of 11.8 percent compared with 18.9 percent for the 100-dma approach. Thus the 100-dma approach beats buy-and-hold with a lower standard deviation: 19.3 versus 28.0 percent. This was coupled with less than half the drawdown levels (maximum loss) experienced by buy-andhold during the worst months. Since 1972, a $1,000 investment in the Nasdaq 100 Index would have accumulated to $28,396 with the buy-and-hold scenario compared with $180,476 for the 100-dma approach without using leverage. Therefore, the 100-dma results were 534 percent better than buy-and-hold with less risk—a great combination.
Using 50 percent leverage on the Nasdaq 100 Index with the 100-dma approach increased the annual return to 23.2 percent, with an ending value of $522,782. For the 100 percent leveraged case (a beta of 2.0 and similar to investing in ProShares leveraged ETFs), the annualized return was 26.9 percent, with an ending value of $1,270,131. In both leveraged cases, the standard deviations were higher than buy-and-hold, but both leveraged strategies had better drawdown numbers—certainly a desirable situation. Again, be aware that investing in instruments with higher betas entails much more risk and that moves in either direction are much deeper and much faster.
Looking at the numbers on a risk-versus-reward basis, the 100-dma timing approach with the Nasdaq Composite Index is superior to buy-and-hold with regard to total return, being out-ofthe- market for 36.4 percent of the time, and with regard to minimizing the losses the bear market slaughters. Also, more aggressive investors may want to use the ProShares leveraged ETFs as their investing vehicle to replicate a leveraged strategy.
In his November 2002 article, Merriman summed up what was learned about using the moving-average approach:
Contrary to [what] many of the critics of timing [say], a timing system can be “wrong” more than 60 percent of the time and still produce a phenomenal increase in return. . . . [T]iming is more effective when it is applied to more volatile assets instead of less volatile ones. . . . [T]iming is even more effective using leverage. And we saw that it always reduces risk by taking investors out of the market at least some of the time.
John Murphy’s 50-dma Nasdaq Composite Index Test
According to John J. Murphy, chief technical analyst at StockCharts.com:
In the 30 years from 1972 to 2002 a “buy-and-hold” strategy reaped a gain of 1,105 percent in the NASDAQ. A simple timing strategy of selling whenever the NASDAQ fell under its 50-dma (and reentering when it rose back above it) reaped a profit of 13,794 percent. In the 10 years from 1993 to 2002, a “buy-and-hold” strategy yielded a NASDAQ profit of 93 percent. By utilizing the “sell discipline” of the 50-day average, that NASDAQ profit jumped to 280 percent.
Mark Courter Backtests 50-, 200-, and 50/200-dma with the S&P 500 Index for Market Buy and Sell Signals
Mark Courter analyzed the performance of using the 200-, 50-, and dual combination 50/200-dma crossovers in comparison with buyand- hold from 1971 through 2009 for market buy and sell signals using a backtesting approach.2 These are exactly the same movingaverage lengths that were described at the beginning of this chapter, except that Courter exclusively used exponential moving averages instead of the simple moving average. This latter average is more sensitive to current data than the simple moving average, which is the reason Courter decided to use it.
In running the analysis, Courter used intraday highs and lows as crossover points above and below the moving-average line for his buy and sell signals because he wanted to avoid whipsaws. Most studies use the end-of-day close or the open on the next trading day for the price component of the backtest. When a sell signal was given, the proceeds were not invested until the next buy signal was generated.
The difficulty with buy-and-hold is the large drawdowns that occur in big bears; for example, the losses in three bear markets 1973–1974 of 48.2 percent, 2000–2002 of 49.15 percent, and 2008 of 56.78 percent—were hard to sit through not knowing when and how long the market needed to recover to the prior highs. Interestingly, the 200-dma strategy lost much less in these three bears, declining 10.4, 11.3, and 15.6 percent, respectively. The 50- dma had worse performance in 2000–2002, losing 31.8 percent, but better performance than the 200-dma in 1973–1974, declining only 7 percent, and in 2008, declining only 15.2 percent.
In conclusion, Courter found the 50/200-dma to be the superior strategy of the three, as Table 9-1 reveals with the best annual return, the lowest standard deviation and the highest ending balance, the second lowest number of trades, and the second highest percent of profitable trades. These results indicate that using the 50/200-dma strategy with different equity investments in conjunction with portfolio diversification can lower an investor’s risk while providing comparable market returns.
TradeStation Nasdaq Moving-Average Tests
In the first edition of this book, I tested the moving-average timing strategy using the Nasdaq Composite Index, first on a weekly basis by using the 25-week moving average (wma) from 1971 through 2002. This is a strategy that an investor with limited time can use. Since Merriman’s research had shown excellent performance using
the Nasdaq Composite Index with the 100-dma, I used the 25-wma, which is equivalent to a 125-dma, which is also very close to McDonald’s 132-dma. I used the TradeStation software to backtest this 25-wma on the Nasdaq Composite Index from February 5, 1971, through December 27, 2002, in the book’s first edition.
Nasdaq Composite Index 25-wma 1971–2002 Test
TradeStation is a powerful trading platform with an extensive backtesting capability. It provides extensive performance reports and charts detailing all aspects of each strategy. The reports show all the pertinent statistics on the test period, including the following:
- Every buy and sell signal delineated with dates and profits or losses
- Daily, weekly, and monthly performance in detail
- Annual net profits by year
- Win/loss ratio statistics
- Graphs of the equity curve
- Monthly net profit, average profit by month, and monthly rolling net profit
The buy and sell rules are shown in the box below. An initial $100,000 was invested at the first signal, and no additional funds were invested. All investments were made on the same day the moving-average crossover signal was given. Dividends, taxes, and margin interest were not included in this analysis.
Nasdaq Composite 25-wma Strategy
Buy signal. When the closing weekly price of the Nasdaq Composite Index pierces the 25-wma from below, a buy signal is given.
Sell signal. When the closing weekly price of the Nasdaq Composite Index drops below the 25-wma, a sell signal is given and a short position is taken.
Table 9-2 shows the key statistics of this simple strategy. Over the total 31-year time frame, buy-and-hold had a return of 968 percent. This did not even come close to the performance of the 25- wma strategy, which returned 4,275 percent. In total dollars, the original $100,000 investment returned a profit of $4,274,870 compared with a profit of $968,400 for buy-and-hold. The annual rate of return for the 25-wma strategy was 13.52 percent.
The ratio of average winning dollars per trade compared with average losing dollars per trade was 4:1, where the average winning trade amount was $157,052 and the average losing trade was $38,648. Only 42 percent of all trades were profitable; however, the profit factor was 2.98. Thus, here again is a case where less than 50 percent of trades were profitable but buy-and-hold was still demolished.
In total, there were 97 trades over the 31 years, averaging just over 3 trades a year, certainly a reasonable number. Twentyfour years had profits and five years had losses. The worst years were 2000 and 1994, with similar losses—of about 16 percent. Two years with back-to-back losses were 1987 and 1988, with about a 7.5 percent average loss per year. And 1977 had a minor loss of 1.4 percent.
Nasdaq Composite Index 25-wma 1971–2010 Test
This original backtest was updated through 2010 in this second edition of this book. Over the nearly 40-year time frame, buy-and-hold returned 1,830 percent, but the 25-wma returned 45,065 percent, more than 25 times more. In total dollars, the strategy had a profit of $4,506,537 compared with $1,830,090 for buy-and-hold. Compared with the prior test period ending in 2002, this strategy has not made much progress from the $4,274,870 gain from eight years ago, whereas buy-and-hold has doubled its return. Table 9-3 provides the complete statistics.
Table 9-4 lists annual returns for this test for 1973–2010, and it shows that the strategy encountered losses in 2004, 2005, 2007, and 2010. These losses held back the progress of this strategy, whereas buy-and-hold did much better, except for the 2008 crash. However, the strategy did have a 17 percent positive return in the big bear market year of 2008. The equity curve (Figure 9-2) for this strategy in the past few years has not been a pretty picture and illustrates the volatility of this strategy.
Nasdaq Composite Index 20-dma 1971–2002 Test
The second backtest in the first edition of this book of the Nasdaq Composite Index used a 20-dma with the same $100,000 starting capital. This is a much faster moving average than the 25-wma and a popular moving average among traders. Using the exact same 31- year test period from February 7, 1971, through December 27, 2002, and the same assumptions and buy and sell rules (except now the 20-dma is substituted for the 25-wma), the results are hard to believe.
The simple 20-dma Nasdaq Composite Index strategy produced a total profit of $46,563,046, which is over 10 times greater than the $4,274,870 realized from the original 25-wma strategy (see Table 9-5). The buy-and-hold strategy for the 20-dma resulted in
total returns of $1,139,260. Thus the 20-dma strategy beat buy-andhold by a ratio of 46:1. The percent of winning trades was only 38 percent. There were 664 trades during the test period, or about 21 per year, compared with only 97 trades, or 3 per year, for the 25- wma strategy. Thus this strategy had 567 more trades than the weekly strategy, resulting in much higher commission costs.
Now look at the equity curve in Figure 9-3 on page 180. The equity curve is a line graph that plots the continuous profit and loss of each trade. The graph indicates a slow but solid start; then, around trade 190, a rise to a peak at trade 350; then a flat period until trade 500; and then an upward surge punctuated by huge downswings after trade 600. Although this equity curve was not a smooth, upward-sloping curve, which would be the ideal scenario, at least it held its own during high-volatility periods and the 2000–2002 bear market, and it ended at its high for the period.
Overall, the 20-dma strategy had only three negative years (–7 percent in 1977, –11 percent in 1993, and –1 percent in 1994) in its 31-year history. In the terrible bear market years of 1973 and 1974, this strategy had a 49 percent and a 4 percent back-to-back positive return. In the 1987 bear market, it gained 46 percent. And in the 2000–2002 bear market years, it gained 4, 32, and 16 percent, respectively.
Nasdaq Composite Index 20-dma 1971–2010 Test
The most recent backtest of the entire period 1971–2010 had a total profit of $45, 814,597 (see Table 9-6). This was actually less than the profit for the strategy ending in 2002. The return on capital was 458,146 percent compared with only 2,179 percent for buy-andhold, a 210-fold increase. There were negative annual returns in 2003 (–15.7 percent), 2006 (–1.42 percent), 2007 (–16.56 percent), and 2009 (–12.11 percent). Interestingly, in the 2008 big bear market, there was a gain of 26.21 percent. All the annual returns are shown in Table 9-7.
Figure 9-4 on page 182 shows the equity curve for this strategy, and you can see the choppy performance during the past few years.
Other Nasdaq Composite Daily Moving-Average Backtests
I also ran a TradeStation optimization analysis of the daily Nasdaq Composite Index moving averages
between 15 and 125 days to determine their profitability. The exact same test period as the other backtests was used: the 31-year period from February 7, 1971, through December 27, 2002, along with the same assumptions. Using an 18- to 22-dma, I found the total returns for the period to be in the $37 million to $56 million range. The 23- to 50-dmas produced returns in the $19 million to $28 million range. By contrast, 100- and 125-dmas had total returns of about $6 million. Thus my testing confirms Colby’s findings that shorter moving averages have higher returns than intermediateand long-term moving averages.
The highest returns for the Nasdaq Composite Index were clearly in the 18- to 22-dma range. The optimal strategy turned out to be the 18-dma, with a total return of $56.02 million and an annualized return of 22.47 percent. This return was almost $10 million better than the 20-dma strategy, which had a 21.74 percent annualized return. Thus, as you can see, the selected Nasdaq Composite Index moving averages provided outstanding returns.
Current Daily Moving-Average Optimization Backtests of the Nasdaq Composite
For this second edition of this book, I also ran another optimization backtest analysis of the daily Nasdaq Composite Index with moving averages in the range of 20 to 30 days. Table 9-8 provides comparative results for 20 to 26 days. Note that the same daily moving average was used for both the buy and sell criteria unless otherwise noted (last two examples in Table 9-8).
Clearly, moving-average lengths greater than 22 days had inferior results to the ones that were shorter in length. Interestingly, dramatically improved results would have been attained using a combination of 21 for the moving average for the buy signal and 22 for the moving average for the sell signal. This combination had a total profit of $65.42 million. However, this observation is known only in hindsight and by using an optimization program, so these optimal moving averages may not be optimal going forward. Nevertheless, both the 21- and 22-dmas are so close to the 20-dma that using the 21/22-dma or even the 20/22-dma combination could be the more profitable approach going forward.
MARKET TIMING USING MOVING AVERAGES
Using moving averages as a market timing strategy is quite simple to execute but often difficult to sit through because the price volatility from year to year can be quite high and unnerving. As the research on moving averages indicates, there are profitable strategies whether you selected the Nasdaq Composite 20/22-dma, the 25-wma, or the combination 50/200-dma. So it is your choice to determine if any of these strategies best fits your risk profile.
Keep in mind that when using shorter-term daily moving averages, you will have many more trades than using longer-term daily or weekly moving averages. This will result in a greater frequency of smaller gains and losses than using longer-term moving averages. Moreover, be careful to assess your risk when using intermediate- and long-term moving averages. With these strategies, you should carefully consider implementing a stop-loss rule to protect your profits so that the investment does not go too far against you when the trend changes. For example, if you are making a profit of 30 percent on your investment, you don’t want to give it all back when the market reverses direction and goes all the way down, crossing the moving average. Thus consider using an 8 to 10 percent trailing stop-limit order to protect those profits. If you wait until the actual moving-average sell signal occurs, then you could be giving back most of your profits. Use common sense in protecting your capital when the market goes against you.
If you decide to use a moving-average strategy, you can go online to any of the charting Web sites mentioned previously and bring up a multiyear chart of the index you select with a selected moving average. For example, let’s say that you choose the Nasdaq Composite Index’s 25-wma strategy. Each Friday afternoon before the market closes or on the weekend, you would bring up the weekly chart to determine if the moving average crossed over or under the index’s price. When a buy crossover occurs, you could purchase the QQQQs (Nasdaq 100) or a bullish Nasdaq fund such as Fidelity’s ONEQ ETF. Then you would wait for the index’s price to decline below the moving average or hit it’s stop-limit order (whichever occurs first) to either sell your position by going into cash or going short the QQQQs or an inverse Nasdaq 100 fund such as a ProShares Short QQQQ (ticker: PSQ).
Moving averages, as a timing tool, have been used for decades by savvy investors. Overall, they perform better than buy-and-hold in bear markets but sometimes worse in bull markets (owing to whipsaws that reduce overall profits), but overall they are a viable strategy. Nevertheless, with a portion of an investor’s money, for example, 20 percent, the moving-average approach, such as the 50/200-dma crossover strategy, certainly has performed well with less risk than buy-and-hold. The more aggressive Nasdaq 25-wma and 20/22-dma are more suited for aggressive investors who want the opportunity for higher returns with more risk.