game theory fails a test

Justin Schwartz jkschw at hotmail.com
Mon Mar 18 10:08:10 PST 2002



>
>[game theory always struck me as a bit of a crock, but I never knew
>quite why - here's one reason - the original paper is at
><http://www.people.virginia.edu/~jg2n/ten&ten.pdf>.]

These results are consistent with similar ones that have been known since the 1950s. Daniel kaneman and Amos Twersky have made a career of doung this ort of experiment. Its significance is very musch up for grabs, however, and there is a huge litearure on it. I do thing that game theory does not describe human behavior with the sort of accuracy that people like Gary Beckerassume. I think that id you get right down to it, probably something like Kanemann & Tverksy' prospect theory (which is much harder to use) may be moreaccurate. All that said, game theory really is very good for important real world results, including the collectiveaction and public goods problems, which are variations on the prisoner's dilemma, and thre sensitivity to starting point of the outcome, which is a collralry of Nash's equilibrium theorem. In sort, the two fundamental results of game theory are powerful, explanatorily valuable, and not "a crock."

Rationally choicishly yours

jks

Without insights
>from behavioral economics and other fields, pure game theory can be a
>beautiful minefield.
>

Quite right too.

jks

_________________________________________________________________ MSN Photos is the easiest way to share and print your photos: http://photos.msn.com/support/worldwide.aspx



More information about the lbo-talk mailing list