In all tests we found the behavior for any particular run does not
necessarily reflect the average behavior of the system. The prices
have a tendency to get ``stuck'' into particular patterns, which
eventually change. The results presented here, therefore, are based on
averages of many runs. While these averages seem very stable, and a
good first step in learning to understand these systems, in the future
we will need to address micro-level issues. So far, we notice that the
micro-level behaviors are much more closely tied, usually in intuitive
ways, to the agents' learning rates 
and exploration rates
. That is, higher rates (for both) lead to more volatile
pricing behavior.