DIRECTIONAL VERSUS ENVIRONMENTAL SIGNAL

Anyone who has ever worked on a project involving the development of any particular indicator, is aware of the fact that "not all divergences are created equal!"  Normally, a divergence between price and an indicator, is a warning sign that price  may be soon changing direction. Eighty to ninety percent of all divergences between price and an intermediate, or, long term indicator, "signal" a potential change in the direction of price. That is the reason,  most people are aware of only this type of signal that is derived thru divergences,  also known as "directional signal." Divergences signify directional signals, only when the "environment" from which the measurements are taken, has remained the same thru-out the time period the indicator is designed to cover. The input values must come from the same type of sample. For example, let's say we are employing a 50 day XYZ indicator, and over the past 20 days, there has been a significant change in the market environment, that means the input values of the indicator come  essentially from two different samples, one that covers the first thirty days, and one that covers the last twenty days. Since the market "has changed" the last twenty days, our indicator is no longer comparing apples to apples. Consequently, there is about to be a distortion, which will appear as a "divergence." That "divergence" will remain in place, until all the data from the previous market environment are filtered out, and are replaced by data that reflect the current environment,  so once again we have a homogenous sample.  

If a divergence signifies a  "directional signal" then on average, price will  conform within a time period less than half the time period the indicator is designed to measure. For example, if a 50 day XYZ indicator starts to diverge today, we should expect price to follow within the next 25 days at most, in the mean time there should not much price movement. So, if we observe a divergence, but price remains unaffected, and more importantly it continues to defy the indicator passed the half point, then we should begin to suspect that our indicator is giving us a much more important signal, it is giving us an "environmental signal." More likely, there has been a significant change in the market's environment, which accounts for the price's ability to maintain course despite the "divergences." Our indicator's inputs are coming from two different samples, that  is why we are getting the divergence! 

When a variety of intermediate/long term  indicators have been diverging with price, for a period longer than half the duration of the period they are measuring, then we need to consider the possibility that the market is undergoing a  "cyclical" or "secular" change. If it is, price will continue to move in its own direction, and the indicators will again begin to confirm price, as soon as, the data from the previous environment are replaced with all current   data. Usually, if a "cyclical" or "secular" change has taken place, in addition to the price being able to stay the course, we will also observe, initially all the indicators to diverge, but as time goes on, we will see that shorter term indicators begin to confirm price while the longer ones continue to diverge. For example, let's say we use the same indicator for two different time frames, a 30 day XYZ indicator, and a 60 day XYZ indicator. Initially both will diverge from price, but if 30 days later, price has held up, the 30 day XYZ indicator begins to confirm price, while the 60 day is still diverging, then more likely the divergence is due to data distortion, because as soon as 30 days worth of latest data were inputed in the 30 day indicator, it started to confirm price.

Many people who are not familiar with mathematical concepts, look at an indicator that is diverging, and because price is not complying, they think that either the indicator has become useless, or, the market is "strange" or "weird!"  In reality, the indicator, may be giving a much more important signal, one of a "cyclical" or "secular" change in the market, opposed to a directional signal, that indicates  simply a change in the direction of price.

In our latest monthly report, I pointed out that for the previous four weeks, none of our indicators had confirmed price, yet price had continued to move on. To me that was a development that needed to be closely scrutinized,  because it may mean that the  market was undergoing  a cyclical or secular environmental change. I pointed out that we still needed more data to confirm such a change, and I was not sure if that would turn out to be case. However, I had seen enough evidence to convince me that such a change  could not be ruled out.  

In an effort to pinpoint where the divergences were coming from, and  to examine them more closely, last week we calculated all of our indicators manually for the  last three months. In other words, instead of feeding the data into our indicators, and getting the output thru the use of a computer, we entered  all the data in the formulas for our indicators  for each day manually, and we calculated the output for each day. The purpose of this monumental exercise was to identify if there was a  point at which some of the independent variables  had changed so dramatically, that would have resulted in changing the current  cyclical, or, secular character of the market.

We came up with several interesting findings, the table below shows the price, breadth and volume figures for the rally off the March lows, and for the rally of the last eight days.

What did we find out? 

NYSE: Between the two rallies we found out that in the latest one that took the SP to 965, total volume contracted by 6%. Net upside volume contracted by  36.7%.  Up volume per Up issue  contracted by 19.5%, while down volume per down issue increased by 23.3%. In March, we had 561,000 shares, for every one issue that went up, and 362,000 shares for every one issue that went down. In May, we had 451,000 shares, for every one issue that went up, and 444,000 shares for every one issue that went down (notice how the spread is closing)  Astonishingly,  the A/D line gained  32.6%,  due to the fact that in the latest rally breadth expanded  on  average by 200  issues on a daily basis.

NASDAQ: 

Between the two rallies we found out that in the latest one  total volume increased by 11.2%. Net upside volume increased  by  1.67%, but total down volume increased by 12.6%. So the increase in total volume was partially offset by the increase in the down volume.   Up volume per Up issue  contracted by 3.5%, while down volume per down issue increased by 16.4%. In March, we had 606,000 shares, for every one issue that went up, and 419,000 shares for every one issue that went down. In May, we had 585,000 shares, for every one issue that went up, and 488,000 shares for every one issue that went down (notice how the spread is closing)  Astonishingly,  the A/D line gained  28.2%,  due to the fact that in the latest rally breadth expanded  on  average by 137  issues on a daily basis.

In conclusion, as the indices marched higher,  total volume contracted, up volume contracted, down volume expanded, and breadth expanded.  In a bull market one would expect breadth to expand as the market is moving higher, which is what this market is doing. On the other hand, you would also expect volume to increase as well, which it has not. In a bear market one would expect volume to decrease as the market is moving higher, which is what this market is doing. In other words, the numbers are telling us that the market  is acting both as a bull, and as a bear! So what is it?  We believe that the contradictory behavior is evidence that the market is trying to make a transition, but the fight is far from over. If the SP breaks above 965, and volume begins to expand, then the bears have lost. One can argue that investors are on the sidelines waiting to see if the SP will negate the Head and Shoulders by  breaking above the neck line,  when that happens, money will come in and volume will increase.  On the other hand, if volume can't expand, and up volume keeps decreasing while down volume keeps increasing, then the  market is about to hit an air pocket, because there is no liquidity to support it.