[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Subject Index][Author Index]

Re: Museum Schizophrenia



In a message dated 3/12/99 7:39:56 PM Eastern Standard Time,
gderkits@lucent.com writes:

<< Agent-based models in the social sciences can model whole
 populations. etc. >>

Every year for years I had at least 3 PhD or Master's students asking me for
lottery data to plug into their rational decision making (economic) models.
They each assumed nothing was known about who buys and why, leaving the field
open.  Soon after the start of my second hour of description about buying
patterns and influencing buying patterns and player demographics, with a good
long side trip into pathological gambling, I noticed a tendency on their part
to grow restless and urgently consider a change of topic.
Before the same happens here, I'd better make the point that models reflect
the assumptions used to produce them.  That's because they are built at least
in part on interpreted results or even speculations.  A major economic
consultant said that the lottery had no secondary economic impact because it
is operated by the government, but it would have secondary economic impacts
(multipliers) if it were privatized.  Calling this a speculation is an example
of courtesy and forbearance.  
Please don't assume that a fully accurate model is easily attained, to say the
least.