The MPT model of Insider Behavior is a 7-factor model - incorporating finely tuned weightings for: Selling and Buying; Transaction size; Transaction date relative to current; Options-related transactions; Historical company behavior; Insider position and; Initial transactions - is not a retro-fitted theoretical model, but has been produced "live" and used by the professional money management community for 18 years. Past performance spreads - between companies ranked best and those ranked worst – have clearly been positive over the period the research has been live.

An interesting corollary to the theory is found in the aggregation of Insider behavior by Industry or Economic Sector. The advantage of cumulative data points amongst aggregation of like-companies provides additional statistical confidence in the behavior observed in this Insider model of robust design. In this approach, Analyst and Market pricing behavior is also aggregated in order to develop a sense of the larger picture.

Probability Plots, and the Informational Hierarchy. It is one thing to have access to information; another to model it in a cohesive, and actionable manner; and yet another to actually “see” the data in a meaningful context. The graphic provided by the MPT Probability Plot allows visualization of not only Market Relative and Self Relative behavior, but of the interaction of a given set of related behavioral inputs as well.

The following graphic illustrates these points. To orient you, the view reveals scores for 4-MPT Behavioral model inputs: Insider (I), Earnings (E), Relative Price Strength (T); and Summary (S) - (the latter combining all models into one).

The highest scores for each model are located at the outside of each model arm and the lowest to the middle. The shaded areas represent the “normal” range for each model. The red dot reveals the current model score for each variable. A score toward the center, white area, indicates out of range behavior in a negative direction. Think of the connections between current values for all four models as creating the red, shaded overlay on the normalized outcomes. Using this view, one is able to instantly ascertain the condition of the company, industry or sector relative to both other companies, and to itself.

Another feature provided by this view, is that of interaction amongst the various behavioral models. In the MPT view, and after working with this concept for 18 years, we have found that the information flow moves from the Insider signal through Earnings, and is then expressed in a market pricing response. The arrow in the graphic illustrates this flow. Another term we use is that of a “Behavioral Market Clock” , with the Insider input located at the starting, 12 O’clock postion. (The behavioral flow need not be through the Earnings/Analyst node and can be expressed directly in terms of price performance.) An adequate Insider "signal" should cause the practitioner to begin focusing on earnings revisions for signs of movement in the same direction implied by Insider behavior. Managers basing their investment decisions on Value or GAARP should be particularly interested in this type of modeling, as it has the potential for improving the timing of longer-term entry and exit points.

Using this approach to the concept of a “Market Clock”, and beginning with Insider change, by the time the model score improvement gets around to the Summary model (S), all models have kicked into gear - followers of Insider behavior, Earnings models (revisions and surprise), and Price Momentum, are all “on-board” - and the share price has accelerated significantly.

In the final analysis, the basis for our work is founded on the principle that in a well-designed model, Corporate Insiders – whether by action or inaction – buying or selling – provide initial clues as to how they perceive the relative attractiveness of their own company shares. Those clues gain cachet by Insider's proximity to corporate developments, training, and the wherewithal to act. They have tended to be correct and serve as primary nodes in our particular Behavioral model – more often than not leading both the analyst community and price direction.