A future project

I'm finishing my master's degree, and the only thing I have left
(other than one exam in Game Theory) is my thesis project. My
current intentions are to complete a project revolving around
prediction markets.

So I discussed this with my advisor, and realised that he doesn't
view PM's as a particularly useful tool. He believes that if you sat
down with an expert (defined as one of those rare people that always
beat the market) and re-created that person's mental model you could
re-create their judgement. My thoughts are 1) it's damn hard to find
that person and, 2) it would be incredibly difficult to develop that
model and that the model is fine-tuned over time.

As I went home and thought about how I could reconcile these ideas, I
had a thought. In the paper "Information Aggregation Mechanisms:
Concept, Design and Implementation for a Sales Forecasting Problem"
by Kay-Yut Chen and Charles R. Plott they described a prediction
market that HP used to forecast sales. The prediction market
regularly beat the HP forecasts; though Chen and Plott didn't
specifically investigate why. Paper is here: http://www.hpl.hp.com/
personal/Kay-Yut_Chen/paper/ms020408.pdf

My dream project would be to re-create the situation where a
prediction market can be directly compared to an "official"
forecast. Then by examining how the players in the market made their
choices and what influenced them I can determine where it is similar
and where it departs from the theory behind the "official" forecast.
Are there just a few experts in the market and the rest are following
the herd? Is everyone aware of the forecast (specifically or
generally) and just making an adjustment to that, or are they
determining a distribution on their own? I think this would be an
interesting project, serve my ends to investigate prediction markets,
and satisfy my advisor.

Of course the problem with this idea is that I need to find a company
that already does sales forecasts (or something similar) and get them
to agree to all this. Clearly an issue, and I may end up going back
to revise my ideas. We'll see.

Cheers for now.