torekp comments on Knightian uncertainty in a Bayesian framework - LessWrong

21 Post author: So8res 24 July 2014 02:31PM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (2)

You are viewing a single comment's thread.

Comment author: torekp 28 July 2014 01:20:49AM *  5 points [-]

If I expect my credence for H to vary wildly then I may delay my decision as long as possible. Furthermore, if the bets are for money (rather than utility) then I'm all for risk aversion.

My utility for money is so close to linear, that for any bet amount I've ever encountered in real life, any non-linearity can be ignored. There's a better reason to be bet-averse: the very fact that the bet is offered is evidence of potential foul play. In the real world, those who offer gambles often turn out to be con artists. As the saying goes, "If you look around the table and don't know who the sucker is, it's you."