My box Q (calling fellow math nerds)

[swiggidy]

So, the previous question was posed as two buttons, but you could rephrase the question. You have $1 mill net worth and there is a button you can press that will give you 50% chance at 100x return and 50% chance at complete loss. Isn't this a question that could be answered with Kelly Criterion? This would be an aggressive strategy and doesn't factor life style differences between 0 and $1 mill, but would still provide an answer.

Further, based on 50% chance you could calculate some multiple where there is not a clear right or wrong answer. I am assuming that $1 mill is vastly greater than anyone's net worth, which is likely fair for 99% of the posters here.

I would do this math but I've been drinking and am lazy.

[wufwugy]

I think scenarios like these are relevant more in areas of psychology even though they can appear to be relevant in probability. One way of seeing this is if you replace the numbers with 1 trillion and 100 trillion; then you clearly see that it's a psycho-social issue and there is no need whatsoever to risk 1T to make 100T

If you have enough of a steady income and personal ambition, then probability can be the dominating factor because you can view this as one part in a series of bets. But without that, the outcomes aren't worth the risk

[BankItDrew]

wuf ruined it 4 me

[wufwugy]

wuf ruined it 4 me

[Penneywize]

It's a really basic problem in finance theory / microeconomics and there is no right or wrong answer; it depends on the type of individual who is asked the question.

I am certain that there isn't any way to say that "X" is unequivocally the right choice.

@wuf, I'm sure you could approach it from the perspective of psychology, but all this really boils down to is risk aversion and utility of money

edit: that bit you talked about with the 1T v 100T is perfectly described by utility of money theory

[rong]

This led to an interesting if not wasted hour of Wikipedia maths links.

[Penneywize]

Risk aversion - Wikipedia, the free encyclopedia

ya

[jackvance]

I was promised some heavy math

[surviva316]

I was promised some heavy math

Some good maths in a somewhat relevant thread:

http://www.flopturnriver.com/pokerfo...fs-184533.html

Basically, speaking to what wuf says, it is a bit of a psychosocial question, but like all things like this, it boils down to math some way or another. It's debatable and even subjective what factors you plug into the equation (psychosocial factors, for example), but in theory there is some perfect set of inputs you could plug in to get a perfectly representative set of outputs.

[swiggidy]

I like the way you're thinking. I was starting with a simplistic case, but you could build a model and factor in things like utility and an individual's risk aversion, etc. You wouldn't come up with "x" is the answer, but maybe you could come up with "px" is the answer where p varies per person.

rong, at least I know this thread was worth while for one person

[Penneywize]

You wouldn't come up with "x" is the answer, but maybe you could come up with "px" is the answer where p varies per person

Yeah you can do that.

I think an interesting application would be, for a given utility function what would be the amount of money needed to be paid to an individual for him to be indifferent between taking the gamble and not taking the gamble (aka certainty equivalence). Then, if the certain amount of money paid is greater than that amount, said individual would choose the riskless option. If it is less, then they would take the gamble.

Alternatively and somewhat less interestingly you could pick a number between the expectation and the certain payout and find out how risk averse an individual needs to be to accept it. It wouldn't be as fun though as you'd end up with something like "hey the second derivate is -2.5" and we can't exactly quantify how much "more risk-averse" that is than -1.

[Penneywize]

Some good maths in a somewhat relevant thread:

http://www.flopturnriver.com/pokerfo...fs-184533.html

So funny how I went off in that thread about risk aversion too. luhl