ChartWatchers

All-in-all, last week was a down week for the major market averages. While the Dow managed to eek out a tiny gain, the other indices fell with the Nasdaq (-2.79%) leading the way. So far this year, the energy-heavy Amex Composite (+5.64%) and the Russell 2000 (+4.75%) are outpacing the other markets with the Nasdaq off 0.38% and the Dow essentially where it began the year.

"Change" appears to be the theme for this week's edition. John and Richard both look at changes that are underway in the sector picture, Carl has indications that change is coming to the Precious Metals sector, and Arthur looks at changes in the Nasdaq's OBV chart. But first, I hope to "change" your approach to scanning...

SCANNING FOR CHANNELS

In the past, I've written articles on scanning for Divergences, Cross-Overs, and P&F Signals. This week I'd like to demonstrate how one goes about scanning for Channels. A Channel is a pattern where a stock has been moving sideways within a given range for a given period of time. In the example below, DELL has been moving sideways between 30 and 40 for over 6 months.

In order to scan for stocks in a channel, you have to decide up front what the upper and lower bound of your channel will be along with minimum the length of time that a stock has to remain within that range. Let's see if we can find other stocks that have been bouncing within the same 30-40 channel that DELL has been in for at least 6 months. We can use the Standard Scan Interface for this (I recommend always using the Standard Scan Interface whenever possible). The "secret" to creating this kind of scan is to use the "Max. High" and "Min. Low" functions. Each one takes a single parameter which is the number of periods (days or weeks) to look back. Here's what the completed scan looks like:

The key lines are in the "Additional Technical Expressions " section. The first line makes sure that we only look for stocks that have not risen above 40 during the past 6 months (= 6 x 4 = 24 weeks = 24 x 5 = 120 days). The second line checks to see that we only find stocks that have not fallen below 30 during those same 6 months.

I can run this scan and then use the CandleGlance technique described in my article "Visual System Development" to verify that the results returned by the scan are correct. Right now (Saturday, April 17th), the scan is returning 63 stocks including DELL (of course), YUM, and even QQQ!

SCANNING FOR CHANNEL BREAKOUTS AND BREAKDOWNS

It's a simple matter to enhance the scan above so that it finds stocks that have just broken out of a channel. Simply add a 1 period offset to the "Max. High" and "Min. Low" criteria lines and then check to see if "today's" close is above/below the limits of the channel. For example, here's a scan that finds all of the stocks that have just moved above the top of the 30-40 channel we've been using:

Right now, only one stock meets that criteria - SSI. Is SpectraSite breaking out? There are several positive developments on its chart:

As always, successful scanning requires you to carefully review these examples, experiment with them until you are comfortable with how they work and then carefully incorporate them into your analysis work. Since scans can be used (and mis-used) in so many different ways, YOU need to invest time and effort in learning all about them first but I guarantee you that it is time and effort well spent. Hopefully, this article (and the previous ones I linked to) will help you get off on the right foot.

DRG/SOX RATIO IS RISING... Earlier in the year I did an analysis of the DRG/SOX ratio as a way to try to measure the mood of the market. The ratio divides the Drug Index (DRG) by the Semiconductor (SOX) Index. The idea is that when investors are confident they buy chips and sell drugs. That pushes the ratio lower. That's what happened during October of 2002 when the market bottomed (see green circle). The downtrend in the ratio continued until last November when it started bouncing (blue circle). Its been trading sideways since then as the market rally has stalled. When investors turn more cautious, they sell chips and buy the more defensive drugs. That pushes the ratio higher. The DRG/SOX ratio is approaching the top of its six-month range and is close to moving above its 200-day moving average for the first time since last spring. The ratio has already broken its eighteen-month down trendline. An upside breakout in the DRG/SOX ratio would, in my opinion, signal a significant shift to a more defensive market mood. The two main reasons for that are rising energy prices and rising interest rates. That also explains why investors are selling rate sensitive stocks and buying energy. None of these rotations are good for the market as a whole.

On Monday we were focused on the pennant consolidation with support at 2038 and resistance at 2080 (gray oval). While these are typically bullish continuation patterns and an upside breakout was expected, it was prudent to wait for confirmation. Instead of the expected, the break came to the downside and the index has moved into corrective mode. A 50-62% retracement of the prior advance (1896 to 2080) would extend to the 1970-1990 area and broken resistance turns into support around 2000. A move below 1970 would be more than just a normal correction and suggest that a bigger decline may be ahead.

On Balance Volume (OBV) was developed by Joe Granville in the 70’s and is as simple as it gets. Volume is added on up days and subtracted on down days. The concern here is the relatively high volume on down days and the relatively low volume on up days over the last few months. With the decline from late January to late March, OBV moved below its December, September and August lows (red arrow) as selling pressure intensified significantly. The late March/early April bounce is a start, but it would take a move above the early April high to get OBV back on the bullish track.

This was an excerpt from the TDT Report, published every Friday. The remainder includes a primer on measuring relative strength, an application of relative strength to the HealthCare SPDR (XLV), a look at gold/XAU with the US Dollar Index, Elliott Waves applied to the S&P 500 and a Model Portfolio update.

DecisionPoint.com tracks net cumulative cash flow of Rydex mutual funds as a way of estimating sentiment in various sectors. The theory is that money 'ought' to follow prices, more or less. In the last several months this indicator has been rather helpful in identifying problematic price moves by the appearance of price/cash divergences.

On the above chart of Rydex Precious Metals Fund, I have highlighted the first divergence with red circles. Note how there was virtually no cash flow supporting the advance into the January price top. An indication that the advance would fail. Next the blue circles show a blowoff move in February. Lots of money moved into the fund, but prices failed to respond positively enough. Again, an indication that the advance would fail.

Now we see a precipitous drop in prices, highlighted by the green circles; however, note that a proportionate amount of cash has not yet fled the sector. To me this indicates an unrealistic optimism, and my conclusion is that prices will have to drop farther in order to increase bearish sentiment to appropriate levels.

This past week brought in "clear view" the under the surface rotation that has been occurring from the technology sector into the healthcare/pharmaceutical sector - and thus we think it important to look at the Pharmaceutical/Semiconductor RATIO. That said, this "repositioning" is extremely important in our overall equity viewpoint; in the past it has coincided with significant shifts in the overall sentiment of stocks to defensive or negative - as a change in risk aversion develops.

This point is quite clear at the Sept-1998 peak, at the Feb-2000 low, and at the Sept-2002 high - each corresponding with a change in trend for the stock market. Hence, we find it extremely important at this point given it is nearly universally thought that stocks correcting to move higher. If this pattern holds, and it still is not clear - then a topping pattern of proportion is developing...that could last several years if past patterns hold.

The bottom line - if you are long - you want to be cautious and consider defensive issues; and if you are inclined to be short...then technology rallies are to be sold, and to be sold aggressively.