REVUE OF A SUCCESSFUL MOVING AVERAGE TECHNIQUE -- 13 AND 34 PERIOD EMAS ALSO WORK ON DAILIES AND MONTHLIES -- PLOT THE EMA SPREAD FOR BETTER TIMING

WEEKLY EMAS STILL WORK... Last July, I reviewed a moving average technique that used weekly exponential moving averages. [I first described this system in October 2005]. I'm revisiting it today because it continues to do remarkably well. And I'd like to suggest expanding its usefulness. The technique is a moving average crossover system. In other words, trading signals are given when the shorter m.a. line crosses the longer. The two moving averages are the 13 and 34 week exponential moving avarages (EMAs). [EMAs are more sensitive than simple moving averages because they give greater weight to more recent price data]. Chart 1 shows the two lines over the last six years for the S&P 500 (absent the price action). Only one crossing (signal) has taken place in those six years. That was a bullish crossing in the spring of 2003 just as the new bull market in stocks was beginning (green circle). Although the two lines converged four times during the four year bull market, the 13 week EMA has never crossed below the 34 EMA. Chart 2 overlays the weekly S&P 500 bars on the two EMA lines for the last two years. The S&P dipped below both lines during the last three market corrections. That doesn't matter. What does matter is that the two moving average lines didn't cross.

Chart 1

Chart 2

THEY ALSO WORK ON DAILY CHARTS... The same two EMAs can be applied to daily charts as well as weekly. And they work just as well. Chart 3 shows what the the 13 and 34 daily EMAs look like for the S&P 500 over the last year. Only two crossings (signals) have taken place during those twelve months. A downside crossing in mid-May (red arrow) was a short-term sell signal. The upside crossing in late July, however, (blue arrow) reversed that to a buy. That buy signal has never been challenged. Chart 4 applies the same signals on the daily bars for the S&P 500. The S&P has gained 175 points (14%) since that bullish crossover.

Chart 3

Chart 4

DAILY EMAS CALL TURNS IN THE NASDAQ ... In case you're wondering if the two daily EMAs work as well on the Nasdaq, they do. Chart 5 shows the 13 and 34 day EMAs over the last year for the Nasdaq. Only two signals are shown. A sell signal in early May (red line) and a buy signal in late August (blue line). That buy signal is still in effect. The Nasdaq Composite is shown in the upper box in Chart 5 for comparison purposes.

Chart 5

ON MONTHLIES TOO ... As is always the case in any trading system, signals on the monthly charts are the last to turn. But they also carry the most influence because they identify the market's major trend. Chart 6 applies the 13 and 34 month EMAs to the S&P 500. A major sell signal was given in early 2001 (red circle) and a buy signal at the start of 2004 (blue circle). That buy signal is still in effect. When dealing with longer-term moving averages, it's helpful to look at the direction of the shorter line. An early sign of a top was given in late 2000 when the 13-month EMA turned down (down arrow). An early sign of a bottom was given in the spring of 2003 when the 13-month EMA turned up (blue arrow). Fast forward to the present. The third blue arrow shows that the 13-month EMA is still rising. [One of the nice things about using a 13 and 34 EMA combination is that you can switch from monthly to weekly to daily charts without changing the numbers. When dealing with simple moving averages, daily values (like 50-days) have to be changed to weekly values (10-weeks).

Chart 6

PLOTTING MA DIFFERENCES ... There's another way to use the 13 and 34 day moving average lines. And that's to plot the difference between them. Chart 7 shows the Russell 2000 Small Cap Index breaking out to a new record high this week. That puts small cap stocks back in synch with rising large caps. Notice that the two EMAs in Chart 7 (13 and 34) came together, but never crossed. They're starting to diverge again. Plotting the difference between the two EMAs tells us something about the trend of a market. The solid black line in Chart 8 is the difference (or spread) between the 13 and 34 day EMAs. The flat horizontal line is the zero line. The crossing over zero last August coincided with a bullish crossing of the two EMAs. During November, however, the EMA spread started to weaken (red arrow). That signalled loss of upside momentum in the RUT. This week, however, the EMA spread bounced off its zero line and turned up (green arrow). That confirms that momentum in small caps has turned positive again. Plotting the spread between the two EMAs allows for even better timing of trades. And you can do it on daily, weekly and monthly charts. Converging EMAs (declining spread) implies some weakening of the trend. Diverging EMAs (rising spread) implies strength. [The EMA spread can be plotted by changing the values for the MACD indicator to 13,34,1].

Chart 7

Chart 8

WEEKLY AND MONTHLY SPREADS STILL RISING ... To finish off this analysis, I'm showing the 13 and 34 EMA spreads for the weekly and monthly charts for the S&P 500. The good news is that both are still rising. The weekly spread in Chart 9 turned up from its zero line last summer (black arrow) and has rising almost uninterruptedly since then. A downturn in that line would be necessary to signal a possible intermediate correction. Chart 10 shows that the monthly EMA spread turned up in early 2003 (black arrow). And it's still rising (see circle). Neither line is cheap by historical standards. That shouldn't be a surprise given the size of the market's advance. As long as the EMA spreads keep rising, however, the market probably will do the same.

Chart 9

Chart 10