Monday, July 12, 2010

Game Theory Musings

Hondo's recent comment got me thinking again (as he usually does). Game theory is extremely useful in decision making but few decision makers actually use it. Too many CEOs got their MBAs for the brand name and networking value and threw away their class notes at the end of each semester.  That leaves the playing field wide open.  I have binders full of my class notes. 

The web gives us plenty of resources for applying game theory to business and financial strategies.  Here's some food for thought:

Are decision trees good enough without game theory?  I've enjoyed using them but incorporating game theory would probably help by minimizing the uncertainty of picking subjective probabilities.

Is game theory useful in scenario analysis?  The easy answer is yes.  Game theory should lend itself easily to forms of scenario analysis that can't be easily stress-tested with tornado diagrams and Monte Carlo simulations.  Mutually exclusive scenarios should be subjected to game theory.  It may even apply to estimates of value at risk when data on trends and volatility is unavailable. 

Constructing a minimax payoff matrix is simple enough for zero-sum games but not every business decision is a win-lose proposition.  Merger arbitrage is sometimes a win-win solution for both targets and acquirers.  An optimal decision for both parties in a proposed merger would require them to work through the Prisoner's Dilemma.  Hmmm . . . sounds like this requires further study.