  |
 |
SPECIAL REPORTS: TECHNICAL EXCELLENCE
In studying the market I have found that indicators work differently at market tops than at market bottoms, which is why I incorporate "neutral" states in all of them. Indicators very rarely travel from buy to sell without stopping for a while at neutral. It is difficult to show neutral states graphically, which is why they do not appear in my charts. However, because indicators often fall into neutral states, buy and sell signals are best when they are fresh. The Russell 2000 Boomerang Indicator is no exception. Just below is the chart I have used for six years without the parameter called "power," which has been dropped to uncomplicate the chart. Over the years the buy signals have been accurate. They are not always perfectly timed, but they have been close, and have markedly outperformed the sell signals. Look at the S&P 500 sell signal in May 2003 (blue circle) -- it was triggered when the light violet line (the oscillator) dropped below -20, and it was very premature. I have discussed the distinction between the oscillator and the trend (orange line). The trend is calculated from the number of stocks that are over extended each day. The more stocks that are overbought the lower it falls, and the higher it climbs the more stocks are oversold. In reviewing the data collected since I first formulated the boomerang concept in 1999, I have found that the trend works best as a sell indicator while the oscillator works best for buys. For ease in comparison I have placed the original chart directly above the new version. The buys are still based on the oscillator, so they are the same as before, but the sells are now based on the trend, instead of the oscillator. I have renamed the oscillator the "buy line" and the trend is now the "sell line."
There are several important differences. First notice that the sell signal (blue circle) moved later by almost 4 months, from April 25 to August 13, 2003 and resulted in a gain of 86 additional S&P 500 points. Notice also that the penetration of -10, which produced the sell signal, lasted just one day and did not appear at all in the Russell 2000 chart below. It came close, but there was no Russell sell signal until March 18 2004. The Russell 2000 and S&P 500 differ in two important ways which have the opposite effect. The fact that there are 2000 rather than 500 stocks in the Russell index makes it more stable, but the fact that the S&P 500 includes big-cap rather than small-cap stocks makes the Russell less stable. Together the two indicators help us apply the Boomerang concept with more reliability than either one can do alone. Keep firmly in mind the fact that these indicators taken together, account for just one of the 29 indicators making up the Summary Index, so they are not to be followed independently by rote.
Here are the charts for both indices going back to the first signals in January 2000. For the S&P 500 the circles call attention to a very early buy signal in June 2002 and to the mid-2003 S&P sell signal that was not mirrored by the Russell 2000, and which is discussed above.
The long term Russell Boomerang chart also has two circles, the one on the right calls attention to the same premature buy signal in Mid-2003 and the one on the left highlights a one-day buy signal that caught a brief rally in October 2000 and was quickly reversed.
MORE, FROM THE JULY 29, 2005 REPORT The Boomerang concept provides me with a way of studying the "chart patterns" of huge numbers of stocks without having to process actual charts -- just as the Goodman Price/Volume Stock Selection System (GPS) does. However, in the case of the Boomerang technique a great deal can be learned from a study of the indices themselves. For example, at the close yesterday 42 stocks in the NASDAQ 100 were in uptrends, 30 were neutral and 28 were declining. Considering the general strength in the market it is revealing to discover that it is not a homogeneous advance.
Let's look at a typical Boomerang Chart. There are 12 lines in the top half of the chart, one line (light blue with markers) is a plot of the NASDAQ 100 since mid-July 2004. The other 11 lines represent moving averages of various lengths. The patterns made by the lines as they move back and forth across each other clearly show where the index has been, and they are helpful to me in projecting where they are heading. My first attempt at developing a notation system was two years ago. At that time I assigned a daily "signature" to each stock or index. While the signatures are useful in providing a snapshot of the relationship between the various moving averages -- for example the NASDAQ 100 currently has a signature of 2222222212, which describes a series of moving averages in which the short ones are at the top and the long are ones at the bottom, but the second from the longest is above the one just shorter than it. Unfortunately I found signatures to be unwieldy to use in studying and comparing hundreds of stocks on a daily basis. The new notation system permits such studies and I have already started to apply it. The plot at the bottom of the chart below displays the value assigned to the NASDAQ 100 each day. A single number between 1 and 1000 provides all the information about the pattern of the moving averages. It is clear that NASDAQ 100 has recovered from its recent brief decline. The three horizontal colored lines work like a traffic signal with green representing buy, orange representing caution and red representing sell -- the NASDAQ 100 is clearly in a "go" state.
Now for the part that has me excited. Wynn Resorts (WYNN), which was bought by the GPS system on July 5 and was selected as a sell at the close yesterday. It was an excellent trade that advanced 16.9% in three weeks. The arrow on the right shows the buy on July 5 -- it was not picked on the first excursion of the Price Volume Indicator above the buy trigger because that happened before July 1, when we updated the system's parameters as we do at each mid-year and year-end. The previous set of parameters had selected it as a buy a few weeks earlier -- a trade that was terminated under the 30-day rule. Because the recent trade was such a clear example, with sharp buy and sell signals , I chose WYNN to study with the new Boomerang notation system -- the results follow.
I indicated the GPS buy below as "PV Buy." It was a buy on July 5 and it coincides with a big advance by the Boomerang chart -- both systems selected the trade at the same time. The sale by the GPS at the close yesterday is also supported by the Boomerang method as you can see by the sharp decline of the oscillator at the bottom of the chart, where the up arrow is pointing. Now compare that up arrow, representing the decline to the caution line, with the up arrow pointing to September 21, in the middle of the chart. At that point the stock was in the same position it is in now, having slowed down after a sharp rally. However, there is a big difference between the the Boomerang chart below and the GPS chart above. Today GPS is saying "sell" and Boomerang is saying "caution" -- in September the GPS was positive when the Boomerang was saying caution, so the urgency to sell did not exist then.
MORE, FROM THE APRIL 15, 2003 REPORT Moving averages are little or no value when a stock in in a trading range. It must be in a trend for them to gain usefulness. Now that you know how to calculate the trending/no trending indicator line, let's turn our attention to moving averages. Most of the individual moving averages that I use can be seen in the chart below. Looking at the right side of the chart, the current date, the lowest one, in green, is a 50-day moving average. The highest line, in pink, is a 3-day moving average. The definition of an overbought Boomerang Stock, is one which has all 10 of its moving averages arranged such that the shortest average is on top and the longest average is at the bottom. An oversold Boomerang Stock is defined in precisely the opposite way -- its shortest average is on the bottom and its longest average is on top. If you look back on the chart to last July, you will see an illustration of that arrangement.
When the market has been falling, and we surmise that there is a high probability that it will rally, we look at stocks that are oversold and when their shortest moving average, the one on the bottom, moves above the next to the shortest, we buy it. On the other hand, we sell an overbought stock when its shortest moving average, which is now on top, drops below the next shortest average. The space in the middle of the lines results from the omission of a few moving averages of intermediate term length, and the size of that space is a reflection of the degree of overextension. The sum of the overextension for all stocks in an average is charted on the Boomerang charts as the power. When the total space is at its maximum, we expect to see the most violent market reversals. Using this system, it is a simple matter to define the present situation of a stock by the relationships between its moving averages. For example, a stock that is overextended to the upside will have its moving averages in perfect alignment, with each one higher than the next. I assign a number to that situation as follows: If a short average is higher than the next longer average, it gets a value of 2. If it is lower than the adjacent moving average it receives the number 1. Therefore, an oversold stock has the number 1111111111 and an overbought stock has the value of 2222222222. An upside Boomerang stock receives a value of 2111111111, and a downside Boomerang stock is assigned the number 1222222222. In each case the single number references the fact that the shortest moving average is not aligned with the others. It suggests a reversal All of the permutations are found when you study enough stocks. For example, a crossover of an oversold stock, like Lowes in late February, will have a value of 2222211111, and so on. MORE, FROM THE MAY 22, 2003 REPORT If you have been looking at the NASDAQ 100 Boomerang Stocks that I have been charting for you in this section, you are probably aware how very volatile they have been. One day they drop 3 or 5 percent and the next they gain a similar amount. It is not unusual to see this at the end of a rally, just before a sell signal is produced. However, once the sell signal has occurred, it is likely that these overextended stocks will be the biggest percentage losers in the early stages of a decline. Here are three more that just turned down yesterday. Be certain to keep in mind that I have made no fundamental analyses of these companies, and I will not be trading them for myself or for my clients. These stocks have "signatures" of 1222222222 as described in my April 15 report. The tickers of the three stocks charted below are BRCM, IVGN and NVDA.
MORE, FROM THE MAY 23, 2003 REPORT The chart below plots the percent return for 23 trading days starting April 16, for stocks with differing configurations of their moving averages. The smallest return during this sharp market advance, was achieved by the group of stocks that were the most overextended -- those with signatures of 2222222222. The largest returns came from the group stocks that had already had a correction. Those in the middle, with two exceptions, were perfectly arranged according to their signatures. In other words, the returns are correlated with the configurations of each group of stocks.
Here is an example of a stock with a 2222222222 signature. Each short term moving average is higher than the next longer average. UCBH Holdings (UCBH) had this configuration on April 16th, the day to which the arrow points. There is a further discussion of "signatures" in the April 15th report.
First Data (FDC), had a different configuration on the 16th of April. Because its 15 day moving average (tan) was out of order, its signature was 2222122222. This was the result of the sharp correction from which it had just recovered. It would seem, from this brief study, that stocks that have recovered from pullbacks are better positioned to advance in a general market rally, than those that are overextended.
Fred Goodman, CFP, is a fee-only Certified Financial Planner based in Los Angeles. You can send him your questions and comments via email at fred@MarketMonograph. The charts and commentary represent what Fred is thinking about the market and thinking of doing for his own account and for accounts he manages. There is no guarantee that you will profit from trading as discussed herein. You may lose money and Fred assumes no responsibility for what you do or do not do with this information. 
|
