Sunday, August 14, 2011

Adoption/Diffusion Model and Market Timing


Over the course of my post, I've used the $USHL5 as a tool to better understand and interpret market breadth. During the course of this past week a series of events transpired that began to interconnect disjointed knowledge resulting in my acknowledging that after months of staring at this chart on a daily basis I had only a perfunctory knowledge of what this data represented and the story it tells. So for this weekend's project I made the commitment to look deeper into exactly what this indicator is and how the underlying mechanics of it operate.

However, before delving in, I've recently been inspired by a number of concepts from The Three Skills of Top Trading by Hank Pruden, specifically the relationship established between the Wyckoff Model of market stages coupled with the use of the adoption/diffusion model.

 
The adoption/diffusion model is described as thus from Wiki:

Diffusion is the process by which a new idea or new product is accepted by the market. The rate of diffusion is the speed that the new idea spreads from one consumer to the next.

Adoption:
Adoption is similar to diffusion except that it deals with the psychological processes an individual goes through, rather than an aggregate market process
Adoption/Diffusion Model Graph
  • Innovators – venturesome, educated, multiple info sources;
  • Early adopters – social leaders, popular, educated;
  • Early majority – deliberate, many informal social contacts;
  • Late majority – skeptical, traditional, lower socio-economic status;
  • Laggards – neighbours and friends are main info sources, fear of debt.
source: http://en.wikipedia.org/wiki/Diffusion_of_innovations


Using the above model, I've roughly broken down the SP weekly chart over the past 3 years into each of these phases to give a pictorial view of the cycles. The three key phases are Innovators, Early Adopters and Laggards. Innovators are the first individuals to adopt and take risks. This is also a critical point where many emerging trends get buried as ideas, concepts, fashion, technology etc... that does not make it into Early Adopters will most likely fail to stick and carry over to the Early Majority. 

SP Weekly Chart


Phase A is where the Innovators come into play. At this stage it is their buying that begins to stabilize a market bottom. Their buying may stick if it carries over into Early Adopters, or the move may fade and the downside continues if it is not. It is this area that also coincides with the phase transition of markets between markdown to accumulation and mark up to distribution.

Phase E is the stage where the public enthusiasm for the market begins to froth and is also the location of distribution to these Laggards before the mark down process begins. The general market participant is greedy when they should be fearful and fearful when they should be greedy and this typical cycle is expressed in Phase A where they will sell to the Innovators and Phase E where as Laggards they will buy from the Innovators. 

Understanding this model of behavior assists in generating greater insights into the Wyckoff model of accumulation/distribution. By the time the fad has basically played out and the Innovators are on to other pursuits, the average person is buying the product at a discount retailer or second hand store. In the market however, the average person is doing the inverse and actually paying a premium for the hand me downs of the Innovators, which is not only counter intuitive, but counter productive too.  The next step is to see how this relates to the $USHL5 indicator.


The USHL5 indicator is defined thus by StockCharts:
The index is computed by counting up all of the US stocks (on any of the three major exchanges) that are making new 52-week highs and subtracting all of the US stocks that are making new 52-week lows.  That gets you the "Daily" version of this index ($USHL). To get the "Weekly" version, you add up the daily values for the past five days and plot that.

The construction of this is simple enough, subtracting new highs from new lows. But it is helpful to dig a little deeper and understand what exactly happens when price makes a new high. On the surface a new 52 week high seems blatantly apparent, price that closes higher then it has over the past year of trading, but in addition academic research indicates that buying new highs can achieve abnormal returns, so there is an underlying market mechanism behind the significance which can result in abnormal returns. Also, stocks that make new highs show demand- an establishment of market of strength- the more new highs, the stronger the market tends to be.

However there is also another significant factor that often gets over looked in the discussion of new highs and that is time factor.. As a stock increases in price over time and hitting new 52 weeks highs, it is perhaps simultaneously, increasing it's 52 week low as well. What this means is that the floor price increases over time. The inverse is also the case as a stock continues to hit new 52 week lows- the 52 week high over time will decrease as well.

The following 3 year chart of the $USHL5 when addressed from this frame of reference in conjunction with the Adoption/Diffusion and Wyckoff model of accumulation/distribution yields valuable clues to interpreting the underlying market structure.

3 year $USHL5
 
The action in from October 2007 to October 2008 is useful in understanding this phenomena in action.  In October of 2008, the cumulative reading bottomed at -16664. This bottom is nearly a year to the date after the market topped out in 2007. With each passing day during this period the 52 week low moved against price action until they crossed and when they did the reading cascaded downward. The underlying mechanics of how this indicator operates can more further refined in this context..

Within a month of this reading, this indicator snapped back 16000 to -547. What can be inferred at this point is that that over the duration of the correction, the 52 week highs dropped towards proximity of the 52 week lows, enough so that minimal buying pressure was sufficient to snap this back 16000. This makes sense since at this point so much has become so beaten down that it does not take much to snap back so quickly.

It is also during this phase that eInnovators begin to make their presence felt. Additionally it is a zone where new leaders present themselves and begin to buck the dominant trend in the market and start making new highs. As indicated on the chart a divergence begins to show up where the index continued to make a lower low, but the $USHL5 did not. With new leaders emerging and early buyers being supported by early adopters, the index in turn moves higher and the USHL5 continued to make new highs.

As with the up cycle, so with the down. At some point this oscillation will continue in the other direction and the flip side of the coin is indicated by the establishment of higher peaks through April 2010, the zenith being 3852 and from there followed a correction after the flash crash that bottomed in August of 2010. Each subsequent peak of the USHL5 thereafter was lower yet the index continued to make new highs. It is also at this point, being two years into a bull cycle that the new lows over time have also increased, so as the floor rises towards the ceiling, the ceiling falling through the floor can happen rapidly as evident by the recent -4345 low that printed.

Using the Wyckoff model, it is evident that the market underwent distribution and is currently in a mark down/redistribution period. Layering the adoption/diffusion model on top, it is clear that Laggards are entering the market and the Innovators are selling their inventory to them and also beginning operations on the bearish side which may resume the decline should Early Adopters begin to pile on which will lead to a period of further markdown. Additionally, the time factor begins to make its presence felt as price begins to drift towards 52 week lows which continue higher with each calendar day dropping off.

Anything can happen in the market and keeping an open mind and preparations to act at a given time and being on constant vigil and alertness for conducive periods should always be the primary focus of any participant. Although anything can happen does not mean however that the probabilities of events should not be weighed and the current evidence indicates that this is a weakening market and the path of least resistance at the moment is to the downside. Should another range establish itself in this area, it will still take time for the damage done to price to firm before further upside continuation is probable.

No comments:

Post a Comment