Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

The quiz Daniel Kahneman wants you to fail

vanityfair.com

Post reply on HN
Print view View original

91–100 of 122 posts

Re: The quiz Daniel Kahneman wants you to fail

#91

Post by ansible »

erichocean wrote: post #75
neonkiwi wrote: post #44

Earlier quoted context omitted.

The reason that $1000 at 90% odds and $900 at 100% odds are used in this example is the expected value is the same in both cases, making the situations 'equivalent'. A 90% chance of losing $9000 has an expected value of -$8100.

The way I read it is this: Do you want to be guaranteed you'll lose $900? Or do you want a 10% chance you'll lose nothing at all, with a 90% chance you'll lose another $100? So given a choice between being (nearly) totally wiped out, or having the chance of not being wiped out, people take the chance of keeping their cash. Makes sense to me.

Yes, I agree. I think 5b is very poorly framed, and doesn't show what it is puported to show in the linked article.

I also agree that the reasoning changes a lot if this is a one-time event vs. a regular occurance.

Re: The quiz Daniel Kahneman wants you to fail

#92

Post by vbtemp »

radu_floricica wrote: post #78
colanderman wrote: post #23

The questions involving "90% chance of $1000 or 100% chance of $900" always bother me. I never understand why economists think that a rational actor would consider them equivalent; they're not , unless you are making that choice many many times. But if I'm given that chance once (which is presumably what most participants assume, since that's not a choice that comes up often in one's life), it's really then a choice…

The book also discusses that. When the problem is _not_ about a life-changing amount, it is better to take the riskier choice when the expected utility is the same. The explanation is long-ish (and honestly, almost above my head - took me awhile to grok it) and involves the sum of all such incidents over a lifetime, and differences in accumulating utility vs accumulating wealth. Maybe someone who read the book more r…

As Kahneman says in the book - every experienced gambler and trader knows that "you win some, you lose some". Although we may not get multiple chances to repeat the same exact gamble, our intuition tends to lead us to minimize risk when the risk is negligible,and to maximize certainty when things are already pretty certain. This is clearly not the optimal approach. By relaxing our personal constraints a little and adjusting our strategy, over the course of a lifetime and the many gambles we take (e.g., starting that web business) they may pay greater dividends than taking our "default" human strategy.

Re: The quiz Daniel Kahneman wants you to fail

#93

Post by vbtemp »

colanderman wrote: post #23

The questions involving "90% chance of $1000 or 100% chance of $900" always bother me. I never understand why economists think that a rational actor would consider them equivalent; they're not , unless you are making that choice many many times. But if I'm given that chance once (which is presumably what most participants assume, since that's not a choice that comes up often in one's life), it's really then a choice…

I think in Kahneman's book (unless I'm recalling incorrectly), the situation wasn't "90% chance of $1000 or 100% chance of $900).

It was more like "90% chance of $1000, or 100% chance of $850" (i.e., something a little less than P(X)*X). That was the whole point), people are willing to pay a premium for certainty - and the contrary (are willing to pay a premium to turn a 0.01% chance into a 0% chance)

Re: The quiz Daniel Kahneman wants you to fail

#94

Post by colanderman »

vbtemp wrote: post #93
colanderman wrote: post #23

The questions involving "90% chance of $1000 or 100% chance of $900" always bother me. I never understand why economists think that a rational actor would consider them equivalent; they're not , unless you are making that choice many many times. But if I'm given that chance once (which is presumably what most participants assume, since that's not a choice that comes up often in one's life), it's really then a choice…

I think in Kahneman's book (unless I'm recalling in correctly), the situation wasn't "90% chance of $1000 or 100% chance of $900). It was more like "90% chance of $1000, or 100% chance of $850" (i.e., something a little less than P(X)*X). That was the whole point), people are willing to pay a premium for certainty - and the contrary (are willing to pay a premium to turn a 0.01% chance into a 0% chance)

Same deal. $850 and $1000 pretty much both equate to "some large sum of money" in my mind.

Now if the sums were $8.50 and $10.00 instead, I'd likely make the more rational choice (90% of $10), because such choices with smaller amounts of money come up far more often in my life: the sample size will be large enough that the mean approaches the expected value.

Re: The quiz Daniel Kahneman wants you to fail

#95

Post by colanderman »

rwallace wrote: post #62
colanderman wrote: post #23

The questions involving "90% chance of $1000 or 100% chance of $900" always bother me. I never understand why economists think that a rational actor would consider them equivalent; they're not , unless you are making that choice many many times. But if I'm given that chance once (which is presumably what most participants assume, since that's not a choice that comes up often in one's life), it's really then a choice…

A thousand dollars is not a life-changing sum. It's well within the range in which the utility of money is linear in the amount. (Even if it's large compared to what you have in your pocket, by the time you spend it, your life will be the same as it is now.) If we were talking about a million dollars, your answer would be right.

A thousand dollars is not a life-changing sum.

There are many many people who would argue with that.

Re: The quiz Daniel Kahneman wants you to fail

#96

Post by colanderman »

2arrs2ells wrote: post #27
colanderman wrote: post #23

The questions involving "90% chance of $1000 or 100% chance of $900" always bother me. I never understand why economists think that a rational actor would consider them equivalent; they're not , unless you are making that choice many many times. But if I'm given that chance once (which is presumably what most participants assume, since that's not a choice that comes up often in one's life), it's really then a choice…

The interesting thing is not that people are risk averse (and thus choose 100% chance of $900), the interesting thing is that people become risk seeking when it comes to losses (and thus choose 90% chance of -$1000). You're creating a bit of a straw man when it comes to behavioral economists and their view of "rational actors" - the whole field is built around the understanding that there is more to economic decision…

Ya, my theory holds for negative sums too -- I'd rather have a chance of not losing [large amount of money] than lose [almost as large amount of money] for sure. A $900 loss could be enough to bankrupt me -- I need to take the 10% chance.

Re: The quiz Daniel Kahneman wants you to fail

#97

Post by colanderman »

Scaevolus wrote: post #79
ssp wrote: post #77

Earlier quoted context omitted.

Distinguishing between risk-averse/neutral/preferring seems like begging the question to me. Couldn't there be an objective answer to which of three behaviors is the most rational in some situation?

Rational towards what end? Maximizing expected gain and minimizing variance are both reasonable metrics.

Right but which metric is reasonable can (and I believe should) change with scale. I'm risk-preferring when it comes to small amounts -- opportunities to make such decisions come up all the time, so I'll be likely to realize the mean. But I'm risk-averse when it comes to large amounts -- I may only get to make one such decision in my life. Better to minimize the variance here.

This not only explains why people play lottery and buy insurance (playing the lottery non-compulsively involves risking only small amounts of money; not having insurance involves risking large amounts of money), but it also explains why those close to retirement should have risk-averse portfolios, while the young should have risk-preferring portfolios (those close to retirement have few "samples" left to take as it were).

Re: The quiz Daniel Kahneman wants you to fail

#98

Post by colanderman »

MengYuanLong wrote: post #39
colanderman wrote: post #23

The questions involving "90% chance of $1000 or 100% chance of $900" always bother me. I never understand why economists think that a rational actor would consider them equivalent; they're not , unless you are making that choice many many times. But if I'm given that chance once (which is presumably what most participants assume, since that's not a choice that comes up often in one's life), it's really then a choice…

Economists do not all think that way. Just as there are many differing perspectives in other sciences (feel free to cringe if you are a mathematician, chemist, etc.), there are a range of schools in economics as well. Though I cannot fully delve into the topic at this time, variance is certainly taken into consideration by many economists. Depending on the situation, (number of betting cycles etc.) variance is very i…

I'd be very interested to see that, thank you.

Re: The quiz Daniel Kahneman wants you to fail

#99

Post by vbtemp »

colanderman wrote: post #94
vbtemp wrote: post #93

Earlier quoted context omitted.

I think in Kahneman's book (unless I'm recalling in correctly), the situation wasn't "90% chance of $1000 or 100% chance of $900). It was more like "90% chance of $1000, or 100% chance of $850" (i.e., something a little less than P(X)*X). That was the whole point), people are willing to pay a premium for certainty - and the contrary (are willing to pay a premium to turn a 0.01% chance into a 0% chance)

Same deal. $850 and $1000 pretty much both equate to "some large sum of money" in my mind. Now if the sums were $8.50 and $10.00 instead, I'd likely make the more rational choice (90% of $10), because such choices with smaller amounts of money come up far more often in my life: the sample size will be large enough that the mean approaches the expected value.

That's the point - you're willing (we all are, usually) to pay a premium for that certainty. In the book he uses all sorts of figures or probabilities (I remember one case, when it was like 99% chance to win one million dollars, or 100% chance to win ${800,000, $600,000, $400,000} -- starts getting a little tricky there, right?)

Re: The quiz Daniel Kahneman wants you to fail

#100

Post by colanderman »

joshu wrote: post #57
colanderman wrote: post #23

The questions involving "90% chance of $1000 or 100% chance of $900" always bother me. I never understand why economists think that a rational actor would consider them equivalent; they're not , unless you are making that choice many many times. But if I'm given that chance once (which is presumably what most participants assume, since that's not a choice that comes up often in one's life), it's really then a choice…

I actually iterated this using mechanical turk a while ago: http://joshua.schachter.org/2008/09/amateur-economist.html People's behavior changes at the dollar price, too. $900 is different from $9 is different from $90000.

Those are some great data points, thanks for doing this research.
Post reply on HN