In a recent radio interview at www.cfrn.net that I did with Mark Faulk, I spoke of a simple demonstration of the problems associated with statistical analysis in basic research methodology and mathematical modeling.
A solid fundamental forensic statistician would tell you that all statistics have three problems, those being the Probability of an Outcome, the Uncertainty in a decision situation, and the “Ignorance” (this is the lack of adequate information) of the observer. While probability and ignorance can both be quantified relatively, it is impossible to similarly describe uncertainty with any precise number.
The SEC said it was going to use a period of time to permit unrestricted short selling on select markets and companies across the Board, based not on the borrow to deliver model, but a carefully designed process called a "Locate". It would be a basic principle of such a study that it be done on a fair and equitable basis, so that the results of effects on companies could be measured against a randomly selected and representative "Control" sample of markets and companies not subject to this manipulation.
This would permit fair and balanced analysis of the impact of short selling on not just market capitalizations, but also on the entire capital formation process for small and medium size companies in particular.
In presentation of the effectiveness of a drug therapy, the FDA is married to a thing called a double-blind, placebo controlled study model. What the SEC has done here doesn't even approach the problem with the slightest third party objectivity necessary for a drug like Motrin to claim to be effective against inflammation.
I can't imagine that the Federal Task Force(s) going after this matter have not already brought people to their team to analyze not only the research methodologies applied here, but also a serious group of mathematicians to analyze the patterns here in the enormous amount of data available.
I have recommended to all that our team must include such skills, where they do not already exist, or need to be complemented. If we have a pattern recognition expert, I haven't spoken with that expert.
With the number of variables influencing this issue, nothing short of a mathematics background in non-linear math, and mathematical quantum physics is needed.
The enemy here is "Gaming" the market much too effectively for them to not be applying such analysis at some level way above the heads of our markets and investments.
Who could these "Gamers" be? That is the KEY question. I assert it is no one you can see in the open.