 |
bostonbubble.com
Boston Bubble - Boston Real Estate Analysis
|
|
SPONSORED LINKS
Advertise on Boston Bubble
Buyer brokers and motivated
sellers, reach potential buyers.
www.bostonbubble.com
YOUR AD HERE
|
|
DISCLAIMER: The information provided on this website and in the
associated forums comes with ABSOLUTELY NO WARRANTY, expressed
or implied. You assume all risk for your own use of the information
provided as the accuracy of the information is in no way guaranteed.
As always, cross check information that you would deem useful against
multiple, reliable, independent resources. The opinions expressed
belong to the individual authors and not necessarily to other parties.
|
| View previous topic :: View next topic |
| Author |
Message |
admin
Site Admin
Joined: 14 Jul 2005
Posts: 1826
Location: Greater Boston
|
 Posted: Fri May 15, 2009 2:07 pm GMT Post subject: |
 |
|
| GenXer wrote: | admin: Lottery's odds are known. The point Taleb is making is that he only bets on rare events for which the odds are not known, and moreover, are underestimated, so he can make a big profit because the underlying securities (i.e. options for example) are undervalued, so investing a set amount and possibly losing a small amount over time, you wait until the aforementioned random event, in which case you profit like a bandit.
|
But the essay you linked to argued that situations with rare events where the odds are known are highly preferable to situations where the odds are unknown. Taleb didn't talk about his own investment strategy in this particular essay (though I do recognize what you're saying as his MO). It was more general and laid out the principles which no doubt underlie his strategy. My point was that his principles could also be applied to the lottery.
| GenXer wrote: | | JCK: Well, the problem is, the odds are still computable for a horse. If there are 20 horses, your chance is 1/20 |
That's not correct as the horses and jockeys are not all of identical ability. It's not a uniform probability distribution.
| GenXer wrote: | | In some games, the house always wins - you don't want to play those games! |
That assumes that your utility curve is always and everywhere convex. What if you have an s-curve in your utility curve and the game would transverse it? The expected change in dollars can be negative at the same time that the expected change in utility is positive. That is, losing $1 can mean effectively nothing to you, but winning enough to never need to work again could be a huge leap in utility.
- admin |
|
| Back to top |
|
 |
JCK
Joined: 15 Feb 2007
Posts: 559
|
| Posted: Fri May 15, 2009 2:19 pm GMT Post subject: |
|
|
| admin wrote: |
That's not correct as the horses and jockeys are not all of identical ability. It's not a uniform probability distribution.
|
Exactly. Some of the horses/jockeys are known to be good, and some are known to be bad. I'm talking about the truly unknown ones. The "best horse" jockey combination often does win as expected, and the really bad ones often lose as expected.
Also the payout odds, while determined by the house, are also greatly influenced by betting patterns. So I wonder if there are optimal solutions, where one could make money, even in though the house is taking a cut, based on the (high level of) irrationality of the other market participants.
My experience is entirely anecdotal, but it did appear to my eye that the unranked horses were often vastly underrated by the odds, and that the favored horse would often be vastly overrated, in some cases with payouts less than 1:1 on a straight bet, which makes no sense. My guess is that a lot of people bet on exactas (pick 1st and 2nd place) and trifecas (pick places 1-3) and those people generally don't include unknowns in their bets, so this irrationally disfavors the unknowns. The high demand for the "sure thing" in those scenarios keeps the payout on certain horses suppressed. |
|
| Back to top |
|
|
GenXer
Joined: 20 Feb 2009
Posts: 703
|
| Posted: Fri May 15, 2009 2:46 pm GMT Post subject: |
|
|
admin: utility is a flawed concept, and is simply a fictitious construct.
Black Swans are by definition events for which you do not know the odds. Where does it say that Taleb prefers events with the known odds?
Lottery odds are known, therefore it is not what Taleb is talking about. The house always wins with the lottery.
Actually, this is why the house always wins in horse racing - too many people mistake past 'success' for skill, and therefore create biased estimates which are far from reality. It is not the underestimation of the other jockeys, but the overestimation of the 'winners'. Heh. Pwned by randomness 
Utility is a non-linear function which in and of itself is useless because it assumes you can model yourself and your psychological reactions, as well as biases and preferences, which is completely not practical.
When your expectation is negative, you dont play the game, because the more you play it, the more chances are that you will end up losing more than you win. This is what I mean. |
|
| Back to top |
|
|
admin
Site Admin
Joined: 14 Jul 2005
Posts: 1826
Location: Greater Boston
|
| Posted: Fri May 15, 2009 3:38 pm GMT Post subject: |
|
|
| GenXer wrote: | admin: utility is a flawed concept, and is simply a fictitious construct.
|
I disagree. I consider it a very useful abstraction for helping me think about how to act rationally.
| GenXer wrote: |
Black Swans are by definition events for which you do not know the odds. Where does it say that Taleb prefers events with the known odds?
|
No they aren't. Black swans are rare events. Quadrants 1 and 2 in Taleb's essay both have known probabilities and both have black swans.
Taleb says that statistical tools are appropriate for decisions in quadrants 1 - 3. This is what I meant when I said that operations in those quadrants are preferable given that they do work well with the tools available. I did not mean that Taleb prefers known probabilities for his own investing.
| GenXer wrote: |
Lottery odds are known, therefore it is not what Taleb is talking about.
|
It's part of quadrant 2.
| GenXer wrote: |
Utility is a non-linear function which in and of itself is useless because it assumes you can model yourself and your psychological reactions, as well as biases and preferences, which is completely not practical.
When your expectation is negative, you dont play the game, because the more you play it, the more chances are that you will end up losing more than you win. This is what I mean. |
I think it's much more flawed to operate under the assumption that $1 would mean as much to you when your net worth is $1M as when your net worth is $1K. You are essentially using a linear utility curve (whether you call it that or not) out of resignation that a more accurate curve would be too complex to construct. Even if you cannot quantify it, it is the expected value of the benefit that matters, not the expected value in terms of dollars.
- admin |
|
| Back to top |
|
|
GenXer
Joined: 20 Feb 2009
Posts: 703
|
| Posted: Fri May 15, 2009 3:49 pm GMT Post subject: |
|
|
Utility is a flawed concept, and has been shown to be flawed because investors are not rational. This is one of the pitfalls of MPT.
Actually, Black Swans are only events that can not be predicted. If you can use tools to predict them, they are by definition not Black Swans.
Anything having to do with market prices is in the 4th quadrant. We know the odds of lotteries, so you can't make money on them (because everybody can do the calculation!)
Oh, I agree that psychology plays a huge role (i.e. 5% of 1M is definitely better than 100% of $10k - I use this all the time to show my clients that as your worth goes up, beating inflation stops being as important as preserving the principal). But still, mathematically, utility is useless! You can think of your own 'utility' so to speak, but it does not apply to everybody out there. |
|
| Back to top |
|
|
admin
Site Admin
Joined: 14 Jul 2005
Posts: 1826
Location: Greater Boston
|
| Posted: Fri May 15, 2009 3:58 pm GMT Post subject: |
|
|
| GenXer wrote: | Utility is a flawed concept, and has been shown to be flawed because investors are not rational. This is one of the pitfalls of MPT.
|
I am not applying it to other investors, I am applying it to myself in an attempt to be rational. Within that context, it's perfectly valid.
| GenXer wrote: | | Actually, Black Swans are only events that can not be predicted. If you can use tools to predict them, they are by definition not Black Swans. |
Then why does Taleb list black swans in quadrant 1, in which he also explicitly includes casinos and games of chance (like the lottery)?
| GenXer wrote: |
Anything having to do with market prices is in the 4th quadrant. We know the odds of lotteries, so you can't make money on them (because everybody can do the calculation!)
|
You can make money, you just probably won't. But again, it's not the expected change in money that matters, it is the expected change in well being.
- admin |
|
| Back to top |
|
|
GenXer
Joined: 20 Feb 2009
Posts: 703
|
| Posted: Fri May 15, 2009 4:11 pm GMT Post subject: |
|
|
Of course. Its a good thing to be aware of your own biases, but everybody is fooled by randomness sometimes.
What he meant to say was that deterministic or easily modeled events are NOT prone to black swans (and are therefore immune). You can win the MA state lottery 2x, but the probability is tiny. Stock market move of 70 SIGMA is not impossible - we may not know its probability, but its definitely a lot larger than our models would suggest. The difference is in predictability. If you are able to predict something and know the odds, the chance of black swan is tiny, etc. |
|
| Back to top |
|
|
|
|
You can post new topics in this forum
You can reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
|
Forum posts are owned by the original posters.
Forum boards are Copyright 2005 - present, bostonbubble.com.
Privacy policy in effect.
Powered by phpBB © 2001, 2005 phpBB Group
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
