GTO-1-06: Strategic Reasoning and the Keynes Beauty Contest Game
This video from Game Theory Online (www.game-theory-class.org) continues from video GTO-1-05 to show how a game theorist would analyze the Keynes Beauty Contest game. It features Matt Jackson (Stanford).
Пікірлер: 9
this is so amazing! thank you for explaining the game with enough detail to really experience it, it helped me :)
Love this set of videos.
so detailed!
dont know if youre still up for this or give answers, but actually in my course at university, the prof explained that 0 is the only Nash-Equib... I'm a bit confused, your explaination is straight forward and understandable...
@tvuspen
9 жыл бұрын
Maximilian Roller If the interval for choices was from 0 to 100, the equilibrium should be at 0. The general term for n iterations (implied in the video) is 100*(2/3)^n. as n approaches infinity (2/3)^n approaches 0 and hence the equilibrium --> 0. However, since the players are restricted to integers on the interval [1,100] the lowest guess will be 1. 2/3 of 1 is 2/3 and the closest integer answer is therefore 1.
So in the case of a smaller game, if I collude another player (to name a larger number), I would have a higher chance to win?
why does optimal choice has to be a no not more than 67? it could be anything less than 100, right?
@aaronbecker7026
Жыл бұрын
Because the winner is the one who guesses 2/3rd the average guess, 67 would be correct only if everyone else guessed 100. Highly improbable since you will never win by guessing the highest number. This is essentially a race-to-the lowest answer regress.
I'll pat myself on the back because when the game was proposed I immediately gave that answer of "1" to myself for it being the most reasonable, heh. Of course, the mechanics of the Nash Equilibrium lend it to being so interesting as you're not always playing such games with reasonable people, thus needing to keep their perspective into account when you make your filtered, quasi-rational, or debatably the most rational, choice. Then one can consider how many filters can pass over their end decision as they can consider what the others will consider because they're considering what you're considering what they're considering. . .ad infinitum. Well, not forever, as average people won't go that far, but it's possible... This shit is so interesting.