Worthwhile intertubez finds

User avatar
JasonL
Posts: 21851
Joined: 05 May 2010, 17:22

Re: Worthwhile intertubez finds

Post by JasonL » 15 Dec 2017, 16:56

Credit is a risk reward financial exchange. Systems that identify real risks to allow for pricing of such are good. Systems that misrepresent risk and cause perverse pricing are bad. I don’t get how social credit scores would be good in this regard but if they help more accurately priced transactions to take place it’s fine I guess.

User avatar
Warren
Posts: 23312
Joined: 26 Apr 2010, 17:03
Location: Goat Rope MO
Contact:

Re: Worthwhile intertubez finds

Post by Warren » 15 Dec 2017, 17:56

I don't understand what is meant by "social credit score". But before you answer that, answer this; Do I want to know?
Gentrification is undocumented immigration for the left-leaning. - Shem

User avatar
Mo
Posts: 22346
Joined: 26 Apr 2010, 17:08

Re: Worthwhile intertubez finds

Post by Mo » 26 Dec 2017, 20:27

There's a lot in this that is sadly true.

They then effectively round to 0 or to ... ere wrong!

I think that journalists would be wise to put the probabilities in ways that people are familiar with. Like if there's a 16% chance of something happening, say it's the same odds as rolling a 7 on a pair of dice. 38%? Drawing a 10 or higher from a deck of cards. 8%? Drawing an Ace. People are familiar that while these are not the most common outcomes, it's not crazy if any of these things happened in a one shot game.
his voice is so soothing, but why do conspiracy nuts always sound like Batman and Robin solving one of Riddler's puzzles out loud? - fod

no one ever yells worldstar when a pet gets fucked up - dhex

User avatar
Aresen
Posts: 13783
Joined: 26 Apr 2010, 20:18
Location: Great White Pacific Northwest

Re: Worthwhile intertubez finds

Post by Aresen » 26 Dec 2017, 22:24

Mo wrote:
26 Dec 2017, 20:27
There's a lot in this that is sadly true.

They then effectively round to 0 or to ... ere wrong!

I think that journalists would be wise to put the probabilities in ways that people are familiar with. Like if there's a 16% chance of something happening, say it's the same odds as rolling a 7 on a pair of dice. 38%? Drawing a 10 or higher from a deck of cards. 8%? Drawing an Ace. People are familiar that while these are not the most common outcomes, it's not crazy if any of these things happened in a one shot game.
Imagine that a forecast giving Candidate X a 10 percent chance included a prominent link, “How X wins.” It would explain how the polling could be off and include a winning map for X. It would all but shout: This really may happen.
IIRC, Nate Silver did exactly that in the 2016 election and a great many of the 'wiser heads' mocked him for saying Trump had a 35% chance of winning.
If Trump supporters wanted a tough guy, why did they elect such a whiny bitch? - Mo

Those who know history are doomed to deja vu. - the innominate one

Most people don't realize Stephen King downplayed the horror that is Maine. - Jennifer

User avatar
Mo
Posts: 22346
Joined: 26 Apr 2010, 17:08

Re: Worthwhile intertubez finds

Post by Mo » 26 Dec 2017, 22:51

Aresen wrote:
26 Dec 2017, 22:24
Mo wrote:
26 Dec 2017, 20:27
There's a lot in this that is sadly true.

They then effectively round to 0 or to ... ere wrong!

I think that journalists would be wise to put the probabilities in ways that people are familiar with. Like if there's a 16% chance of something happening, say it's the same odds as rolling a 7 on a pair of dice. 38%? Drawing a 10 or higher from a deck of cards. 8%? Drawing an Ace. People are familiar that while these are not the most common outcomes, it's not crazy if any of these things happened in a one shot game.
Imagine that a forecast giving Candidate X a 10 percent chance included a prominent link, “How X wins.” It would explain how the polling could be off and include a winning map for X. It would all but shout: This really may happen.
IIRC, Nate Silver did exactly that in the 2016 election and a great many of the 'wiser heads' mocked him for saying Trump had a 35% chance of winning.
But even if it was only a 16% chance or a 6% chance doesn't mean that Trump winning proves the model wrong. People roll 7 or 3 on dice all the time. Though maybe the craps player in me has internalized it more. The way to test a model is to see how often the upset happens. How many times did a team have a 99.6% chance of losing and end up winning? If it was one out of 275, it's a pretty good model, if it was one out of 2000, it's a pretty bad model.
his voice is so soothing, but why do conspiracy nuts always sound like Batman and Robin solving one of Riddler's puzzles out loud? - fod

no one ever yells worldstar when a pet gets fucked up - dhex

User avatar
nicole
Posts: 8493
Joined: 12 Jan 2013, 16:28

Re: Worthwhile intertubez finds

Post by nicole » 27 Dec 2017, 08:09

Who has a better idea of how likely it is to roll a pair of 7s than they do of what share 16 is of 100?

Anyway I chose not to read the column because when I saw a tweet about it I was like “yep that’s why people are fucking intolerable.” I was told this would stop after I escaped public school, dammit!
"Fucking qualia." -Hugh Akston

"This is why I carry a shoehorn.” -jadagul

User avatar
Mo
Posts: 22346
Joined: 26 Apr 2010, 17:08

Re: Worthwhile intertubez finds

Post by Mo » 27 Dec 2017, 08:18

nicole wrote:
27 Dec 2017, 08:09
Who has a better idea of how likely it is to roll a pair of 7s than they do of what share 16 is of 100?
People who have played any board game. It becomes less abstract.
his voice is so soothing, but why do conspiracy nuts always sound like Batman and Robin solving one of Riddler's puzzles out loud? - fod

no one ever yells worldstar when a pet gets fucked up - dhex

User avatar
nicole
Posts: 8493
Joined: 12 Jan 2013, 16:28

Re: Worthwhile intertubez finds

Post by nicole » 27 Dec 2017, 08:57

Mo wrote:
27 Dec 2017, 08:18
nicole wrote:
27 Dec 2017, 08:09
Who has a better idea of how likely it is to roll a pair of 7s than they do of what share 16 is of 100?
People who have played any board game. It becomes less abstract.
Yeah that is the opposite of me and deeply weird.

I mean you can move the 16 around in your mind and see how it fits into the 100. The dice are just like, not as real as that.
"Fucking qualia." -Hugh Akston

"This is why I carry a shoehorn.” -jadagul

User avatar
Mo
Posts: 22346
Joined: 26 Apr 2010, 17:08

Re: Worthwhile intertubez finds

Post by Mo » 27 Dec 2017, 09:10

nicole wrote:
27 Dec 2017, 08:57
Mo wrote:
27 Dec 2017, 08:18
nicole wrote:
27 Dec 2017, 08:09
Who has a better idea of how likely it is to roll a pair of 7s than they do of what share 16 is of 100?
People who have played any board game. It becomes less abstract.
Yeah that is the opposite of me and deeply weird.

I mean you can move the 16 around in your mind and see how it fits into the 100. The dice are just like, not as real as that.
But It's not that they can see how frequent it is, but people can think of playing Monopoly and thinking, "Come one 7," and not think it's that uncommon (and also that it's the most common result).
his voice is so soothing, but why do conspiracy nuts always sound like Batman and Robin solving one of Riddler's puzzles out loud? - fod

no one ever yells worldstar when a pet gets fucked up - dhex

User avatar
Highway
Posts: 12636
Joined: 12 May 2011, 00:22
Location: the Electric Ocean

Re: Worthwhile intertubez finds

Post by Highway » 27 Dec 2017, 09:19

I dunno that that would change the thinking, tho. People would still think "Well, that never happens when I want it to" and other pessimistic things. I think it's more the way people are that they compress everything to never, 50-50, and always, especially when that's what they want to think.
"Sharks do not go around challenging people to games of chance like dojo breakers."

User avatar
JasonL
Posts: 21851
Joined: 05 May 2010, 17:22

Re: Worthwhile intertubez finds

Post by JasonL » 27 Dec 2017, 09:39

You can't even convince people how streaky a run of 100 coin tosses is. The people who most frequently use dice and cards are the effing worst at contextualizing probabilities. They simply engage higher reasoning to tell themselves a more complicated story about how this time the streak is real or whatnot. We spend a lot of time trying to get people to understand long risk and its freaking hard. Their investment decisions were "right" if they went up and wrong if they went down. Well, no. That's not how that works. Same problem dude is talking about.

User avatar
Warren
Posts: 23312
Joined: 26 Apr 2010, 17:03
Location: Goat Rope MO
Contact:

Re: Worthwhile intertubez finds

Post by Warren » 27 Dec 2017, 09:45

JasonL wrote:
27 Dec 2017, 09:39
You can't even convince people how streaky a run of 100 coin tosses is. The people who most frequently use dice and cards are the effing worst at contextualizing probabilities. They simply engage higher reasoning to tell themselves a more complicated story about how this time the streak is real or whatnot. We spend a lot of time trying to get people to understand long risk and its freaking hard. Their investment decisions were "right" if they went up and wrong if they went down. Well, no. That's not how that works. Same problem dude is talking about.
Except that whether a price goes up or down depends entirely on human activity which often defies statistical probabilities. Economics is pseudo science.
Gentrification is undocumented immigration for the left-leaning. - Shem

User avatar
nicole
Posts: 8493
Joined: 12 Jan 2013, 16:28

Re: Worthwhile intertubez finds

Post by nicole » 27 Dec 2017, 09:56

Mo wrote:
27 Dec 2017, 09:10
nicole wrote:
27 Dec 2017, 08:57
Mo wrote:
27 Dec 2017, 08:18
nicole wrote:
27 Dec 2017, 08:09
Who has a better idea of how likely it is to roll a pair of 7s than they do of what share 16 is of 100?
People who have played any board game. It becomes less abstract.
Yeah that is the opposite of me and deeply weird.

I mean you can move the 16 around in your mind and see how it fits into the 100. The dice are just like, not as real as that.
But It's not that they can see how frequent it is, but people can think of playing Monopoly and thinking, "Come one 7," and not think it's that uncommon (and also that it's the most common result).
I mean, which is more likely to be accurate, someone's impression, or the math? I mean people are completely stupid about how often things happen. If you asked me how often something happens, I'd have to really think about it and calculate the probability before I told you because I know otherwise I'm full of shit because of biases.
"Fucking qualia." -Hugh Akston

"This is why I carry a shoehorn.” -jadagul

User avatar
JasonL
Posts: 21851
Joined: 05 May 2010, 17:22

Re: Worthwhile intertubez finds

Post by JasonL » 27 Dec 2017, 10:04

Warren wrote:
27 Dec 2017, 09:45
JasonL wrote:
27 Dec 2017, 09:39
You can't even convince people how streaky a run of 100 coin tosses is. The people who most frequently use dice and cards are the effing worst at contextualizing probabilities. They simply engage higher reasoning to tell themselves a more complicated story about how this time the streak is real or whatnot. We spend a lot of time trying to get people to understand long risk and its freaking hard. Their investment decisions were "right" if they went up and wrong if they went down. Well, no. That's not how that works. Same problem dude is talking about.
Except that whether a price goes up or down depends entirely on human activity which often defies statistical probabilities. Economics is pseudo science.
If you think a single asset holder riding 12 months of high returns has comparable risk to a diversified asset holder over long time horizons because something something human activity pseudoscience, you are mistaken.

User avatar
Mo
Posts: 22346
Joined: 26 Apr 2010, 17:08

Re: Worthwhile intertubez finds

Post by Mo » 27 Dec 2017, 10:14

nicole wrote:
27 Dec 2017, 09:56
Mo wrote:
27 Dec 2017, 09:10
nicole wrote:
27 Dec 2017, 08:57
Mo wrote:
27 Dec 2017, 08:18
nicole wrote:
27 Dec 2017, 08:09
Who has a better idea of how likely it is to roll a pair of 7s than they do of what share 16 is of 100?
People who have played any board game. It becomes less abstract.
Yeah that is the opposite of me and deeply weird.

I mean you can move the 16 around in your mind and see how it fits into the 100. The dice are just like, not as real as that.
But It's not that they can see how frequent it is, but people can think of playing Monopoly and thinking, "Come one 7," and not think it's that uncommon (and also that it's the most common result).
I mean, which is more likely to be accurate, someone's impression, or the math? I mean people are completely stupid about how often things happen. If you asked me how often something happens, I'd have to really think about it and calculate the probability before I told you because I know otherwise I'm full of shit because of biases.
I'm not saying that it's inaccurate, but anecdotally, I remember talking to people that were psyched that Hillary had an 84% chance of winning and their faces/confidence would change when I said, "That means Trump has the same chance of winning as rolling a seven on a pair of dice." They may not be accurate, but people thing of 84% as "definitely going to happen," but think of rolling a seven as "something that could happen."
his voice is so soothing, but why do conspiracy nuts always sound like Batman and Robin solving one of Riddler's puzzles out loud? - fod

no one ever yells worldstar when a pet gets fucked up - dhex

User avatar
Warren
Posts: 23312
Joined: 26 Apr 2010, 17:03
Location: Goat Rope MO
Contact:

Re: Worthwhile intertubez finds

Post by Warren » 27 Dec 2017, 10:18

JasonL wrote:
27 Dec 2017, 10:04
Warren wrote:
27 Dec 2017, 09:45
JasonL wrote:
27 Dec 2017, 09:39
You can't even convince people how streaky a run of 100 coin tosses is. The people who most frequently use dice and cards are the effing worst at contextualizing probabilities. They simply engage higher reasoning to tell themselves a more complicated story about how this time the streak is real or whatnot. We spend a lot of time trying to get people to understand long risk and its freaking hard. Their investment decisions were "right" if they went up and wrong if they went down. Well, no. That's not how that works. Same problem dude is talking about.
Except that whether a price goes up or down depends entirely on human activity which often defies statistical probabilities. Economics is pseudo science.
If you think a single asset holder riding 12 months of high returns has comparable risk to a diversified asset holder over long time horizons because something something human activity pseudoscience, you are mistaken.
No. I am not saying that.

I'm saying that human behavior is not well modeled by statistics. We use statistics because it's the only tool we have. You need to manage risk when making investment decisions, so what else are you going to do? It's just the best we can do but black swan events happen. And I'm not at all convinced that there's a good model to manage the risk of black swan events.

Contrast with say Brownian motion. You have a tiny dust mote floating in a glass of pure water. The dust mote moves in a herky jerky fashion. How can we predict it's movement. Well, we have a molecular theory of liquids that says the mote is being buffeted by water molecules, and if we measure the velocity of all the molecules we can calculate the results of each collision etc. But the number of molecules is far too large for any such analysis. Fortunately the law of large numbers allows us to model the motion based on random probability, and it's a good model. If, for instance you dissolve a sugar cube in water, a sugar molecule can precipitate out of solution and form a crystal that grows. But what never happens is that a new sugar cube is formed. With people, sometimes they rebuild the cube.
Last edited by Warren on 27 Dec 2017, 10:19, edited 1 time in total.
Gentrification is undocumented immigration for the left-leaning. - Shem

User avatar
Aresen
Posts: 13783
Joined: 26 Apr 2010, 20:18
Location: Great White Pacific Northwest

Re: Worthwhile intertubez finds

Post by Aresen » 27 Dec 2017, 10:19

Mo wrote:
27 Dec 2017, 09:10
nicole wrote:
27 Dec 2017, 08:57
Mo wrote:
27 Dec 2017, 08:18
nicole wrote:
27 Dec 2017, 08:09
Who has a better idea of how likely it is to roll a pair of 7s than they do of what share 16 is of 100?
People who have played any board game. It becomes less abstract.
Yeah that is the opposite of me and deeply weird.

I mean you can move the 16 around in your mind and see how it fits into the 100. The dice are just like, not as real as that.
But It's not that they can see how frequent it is, but people can think of playing Monopoly and thinking, "Come one 7," and not think it's that uncommon (and also that it's the most common result).
When you have hotels on Boardwalk and Park Place and the other players never land on them, you begin to suspect the dice are rigged. ;)
If Trump supporters wanted a tough guy, why did they elect such a whiny bitch? - Mo

Those who know history are doomed to deja vu. - the innominate one

Most people don't realize Stephen King downplayed the horror that is Maine. - Jennifer

User avatar
Aresen
Posts: 13783
Joined: 26 Apr 2010, 20:18
Location: Great White Pacific Northwest

Re: Worthwhile intertubez finds

Post by Aresen » 27 Dec 2017, 10:26

Warren wrote:
27 Dec 2017, 10:18
I'm saying that human behavior is not well modeled by statistics. We use statistics because it's the only tool we have. You need to manage risk when making investment decisions, so what else are you going to do? It's just the best we can do but black swan events happen. And I'm not at all convinced that there's a good model to manage the risk of black swan events.
My first thought on Taleb's "Black Swan" investing was that he was betting on one rare event that had a huge payoff but incurring costs for performing many more transactions, which would totally negate the huge payout.
Warren wrote:
27 Dec 2017, 10:18
With people, sometimes they rebuild the cube.
This has also been observed in Borg.
If Trump supporters wanted a tough guy, why did they elect such a whiny bitch? - Mo

Those who know history are doomed to deja vu. - the innominate one

Most people don't realize Stephen King downplayed the horror that is Maine. - Jennifer

User avatar
JasonL
Posts: 21851
Joined: 05 May 2010, 17:22

Re: Worthwhile intertubez finds

Post by JasonL » 27 Dec 2017, 10:36

We definitely don't have a good way to model black swan events. They are by definition too rare to be modeled. Similarly, there's not a useful model for election probabilities to account for a meteor hitting nyc.

Regarding Taleb's robustness angle on black swan events, I have some sympathies and a lot more concerns. He wants you to arrange your life so that you are robust against the most catastrophic events at the cost of not getting anywhere close to median returns that are available in 99.999999% of cases. He's a bit like an investment prepper living in a bomb shelter. He has a point but, as always, he goes 100 miles too far. The reasonable lesson would seem to me to be something like "take normal portfolio theory and apply an additional liquidity buffer in cash or similar on top of what models assuming normal distributions would suggest". Just hold more cash for off events.

Looking back on 2008 - taleb's model would have been completely crushed by the recovery of someone who just held a blend per normal portfolio theory. This is not investment advice.

User avatar
Mo
Posts: 22346
Joined: 26 Apr 2010, 17:08

Re: Worthwhile intertubez finds

Post by Mo » 27 Dec 2017, 15:41

Related
Image
his voice is so soothing, but why do conspiracy nuts always sound like Batman and Robin solving one of Riddler's puzzles out loud? - fod

no one ever yells worldstar when a pet gets fucked up - dhex

User avatar
dhex
Posts: 15090
Joined: 05 May 2010, 16:05
Location: 'murica

Re: Worthwhile intertubez finds

Post by dhex » 29 Dec 2017, 22:31

https://www.bizjournals.com/sanjose/new ... juana.html

man culture has really turned a corner.
"I do wear my New Balance tennis shoes when I'm wearing cargo shorts, though, because truth in advertising." - lunch

User avatar
Eric the .5b
Posts: 11688
Joined: 26 Apr 2010, 16:29

Re: Worthwhile intertubez finds

Post by Eric the .5b » 30 Dec 2017, 06:11

They've been doing more-or-less those meals for a few years, now. But yeah, being so explicit is a thing.
"Better that ten guilty persons escape than that one innocent suffer."
"Cyberpunk never really gave the government enough credit for their ability to secure a favorable prenup during the Corporate-State wedding." - Shem

User avatar
nicole
Posts: 8493
Joined: 12 Jan 2013, 16:28

Re: Worthwhile intertubez finds

Post by nicole » 03 Jan 2018, 13:03

Warren wrote:
15 Dec 2017, 17:56
I don't understand what is meant by "social credit score". But before you answer that, answer this; Do I want to know?
"Fucking qualia." -Hugh Akston

"This is why I carry a shoehorn.” -jadagul

User avatar
Andrew
Posts: 5935
Joined: 26 Apr 2010, 21:52
Location: Vale of Eternal Fire

Re: Worthwhile intertubez finds

Post by Andrew » 03 Jan 2018, 13:05

Don't fight it son. Confess quickly! If you hold out too long you could jeopardize your credit rating.
We live in the fucked age. Get used to it. - dhex

holy shit there will never be an end until the sweet release of death (as dictated by the death panels, natch) - lunch

User avatar
Shem
Posts: 6880
Joined: 27 Apr 2010, 00:27

Re: Worthwhile intertubez finds

Post by Shem » 07 Jan 2018, 03:00

Image
"VOTE SHEMOCRACY! You will only have to do it once!" -Loyalty Officer Aresen

Post Reply

Who is online

Users browsing this forum: No registered users and 4 guests