EPTUS Politics Mega-thread - Page 118
| Forum Index > Closed |
Read the rules in the OP before posting, please. In order to ensure that this thread continues to meet TL standards and follows the proper guidelines, we will be enforcing the rules in the OP more strictly. Be sure to give them a re-read to refresh your memory! The vast majority of you are contributing in a healthy way, keep it up! NOTE: When providing a source, explain why you feel it is relevant and what purpose it adds to the discussion if it's not obvious. Also take note that unsubstantiated tweets/posts meant only to rekindle old arguments can result in a mod action. | ||
|
Sermokala
United States13743 Posts
| ||
|
sam!zdat
United States5559 Posts
| ||
Souma
2nd Worst City in CA8938 Posts
On February 20 2013 10:27 Sermokala wrote: Ronald reagen only getting a 44 on that scale is silly. Its fucking morning in america and we're gona win this cold war right now goddammit. He instituted an age of cutting taxes and cutting spending by massively raising spending and raising taxes. I would dismiss Reagan's score in any case. 'Congressional Voting Record' is the best indicator, followed by, perhaps, Public Statements. | ||
|
{CC}StealthBlue
United States41117 Posts
Two new polls out today show that support for marriage equality and rejection of the so-called Defense of Marriage Act (DOMA) are on the rise. President Obama hasn’t said whether same-sex couples have a constitutional right to marry. Yet an overwhelming majority of Americans believes that they do. According to a poll conducted for the Respect for Marriage Coalition, “Three-quarters of voters (75%) believe that [same-sex marriage] is a Constitutional right.” This is up four points since 2011. Also, such support “spans across party lines.” Democrats: 91 percent Independents: 75 percent Republican: 56 percent Meanwhile, an overwhelming majority of Americans (83 percent), “regardless of their personal opinion on the issue,” believes that same-sex marriage will be legal nationally “in the next five to 10 years.” The number drops to 77 percent for those who believe marriage equality will go national “in the next couple of years.” The only way marriage equality can go national is if DOMA is repealed either by a ruling of the Supreme Court or by Congress. What makes DOMA especially egregious is Section 3, which defines “marriage” as between one man and one woman and “spouse” as someone of the opposite gender. Because of this, federal law does not recognize legally married same-sex couples. Thus, they are denied Social Security survivor benefits among the more than 1,100 other rights and benefits enjoyed by heterosexual married couples. Source | ||
|
paralleluniverse
4065 Posts
On February 20 2013 10:15 Rassy wrote: Thx, interesting post and i have to agree. The underestimating of the force of correlation seems a verry good explanation of why risk managers ended up in such a bad spot. Its indeed a bit off topic, i just responded to the 3 charts shown and if have to choose between the mandlebrot and the bellcurve,then i go with the bellcurve and i still think that thats the most accurare respresentation (though of course not as accurate as the bellcurve of tossing head or tails 100 times gives) It was merely a remark on the side. Had one question about the post you made before this: "It's possible that Alesina is correct, that tax cuts are less contractionary than spending cuts. But one reason to show some doubt is that if this were true" I asume you mean tax increases (or the ending of tax cuts) since tax cuts are not contractionary at all, tax increases are. Or am i now missing something realy simply or misreading you? That's right. I've fixed some typos in that post. | ||
|
paralleluniverse
4065 Posts
On February 20 2013 07:55 sam!zdat wrote: That's MY point! No. There are also unknown-unknowns. You must always acknowledge that there are unknown-unknowns, and that is what your derivatives can never do. And don't even get me started on the unknown-knowns. You can use the normal distribution and then add in correlation, fine. But here you should just go read Taleb's book, he knows far more about it than I. Oh, I'm well aware of this. They've taken over because they give shallow normative grounding based on flimsy mathematics to an ideologically predetermined conclusion. Of course I didn't show anything pre-2007, it isn't my job. don't be ridiculous. I don't have to be a cassandra in order to be angry at the incompetence of the people who rule my world. My point is that what they claimed to have accomplished was prima facie absurd and everybody should have known that. you can hand out as many bullshit "Nobel Prize in Economics" as you want, that doesn't make it a real academic discipline You never know what the worst case scenario is. The worst case scenario is when your worst case scenario analysis turns out to be wrong, and you can never, never, never, never rule out that possibility. edit: and the so-called "stress test" just took the numbers from the last crisis and tried them again. talk about "trying to model long-tail events from history", as you so blithely accused me of above. This stress test was useless because it just took the number from the last crisis as the number to test. This same stress test would not have revealed the flaws in any of the preceding crises, because they were all bigger than the ones that came before them. So the stress test would have revealed nothing. bullshit. anyway, i've been following the finance class on yale open courses - I'm not totally ignorant of the subject. I get the idea. Financial math is not "flimsy mathematics". In fact, it's some of the most complicated and heavily research math out there. You can say that the models don't fit with reality, when the assumptions don't hold, but that doesn't make it flimsy, it just means that the real world is more complicated, and we don't completely know how to model it. Which is one of the reasons why there's so much research in financial math. I didn't say that a stress test is seeing what happens in literally the worse case scenario. I said it's like worse case scenario analysis in the sense that you test that your firm can survive really bad conditions. If I recall correctly, the stress tests that were run, showed that the banks could survive worse than the financial crisis. You complain that the stress tests were worthless because they assumed a crisis of the same size. But that's not useless. It provides very useful information. It reveals that these banks could survive a crisis the size of the recent one, a once in a generation event. Then you say that crises are bigger than the ones that came before. But this is simply not true, the global financial crisis was not as bad as the great depression. Even if these stress tests were rigged, that's not a criticism of the concept of scenario analysis/stress testing. The idea of doing a stress test, to show, say, that even in situations worse than the financial crisis, the firm can survive is one of the most effective ways of managing tail risk. Of course, it's impossible and impractical to make sure the firm can survive no matter what. You say that we don't know what the next tail event is. But that doesn't invalidate the idea of stress testing. Just assume various semi-plausible catastrophic events. Suppose an asteroid destroys the US, this is a tail event, it is unpredictable, large, a 6-sigma event. No sensible stress test requires that your firm is able to survive such an event. You can "never, never, never, never" rule out this possibility. But that doesn't mean you have to cover for it. | ||
|
paralleluniverse
4065 Posts
Insurance companies deal with catastrophes all the time, (e.g. the Japanese earthquake and tsunami a couple of years ago). Why don't they go bust? No, AIG doesn't count, because it was the investment division, gambling with financial instruments, that blew up, not their insurance business. The main reason is because insurance companies know how to manage tail risk. Stress testing is one method. Another method is reinsurance, i.e. the insurance company buys insurance from another insurance company. The analogous concept in finance is using derivatives to hedge your position. Another is catastrophe bonds. The analogous concept in finance is a MBS. So why hasn't reinsurance and cat bonds destroyed insurance, the way that derivatives and MBSs have destroy finance? Because natural catastrophes, of the sort that insurance companies deal with aren't correlated, whereas everything is correlated in finance. A tsunami in Japan doesn't cause a hurricane in New York and a volcano eruption in Italy, whereas a large bankruptcy causes contagion to flow throughout the financial system. So this proves my point: it's not tail risk that was the main problem, it's correlation. It's possible to successfully deal with tail risk when things aren't correlated, just look at the insurance companies. | ||
|
paralleluniverse
4065 Posts
On February 20 2013 10:27 sam!zdat wrote: ugh, the mandelbrot set was not supposed to be a "representation" of the financial data. The point is just that markets have fractal structure, not gaussian structure. you can use a gaussian model as a tool, but you have to understand that it's fundamentally different than the thing you're trying to model, and the more complex your models get based on this fundamental difference, the more you iterate the error until everything falls apart. So we should use gaussian models, but we should not build such complex things out of them, because we do not understand what we are doing when we do that. If bankers were the people who suffered from their mistakes, I wouldn't care as much. These sentences literally make no sense. It's imprecise, using invented terms, and simply doesn't make sense. For example, what is "Gaussian structure" and why is it not "fractal structure". What's specifically is Gaussian? If we assume that stock market returns have a Gaussian distribution, than the path of stock prices follows a Brownian motion. And Brownian motion is a fractal, i.e. if you keep zooming in, it looks like itself. This is the famous self-similarity property of Brownian motion: http://en.wikipedia.org/wiki/File:Wiener_process_animated.gif Mandelbrot introduced fractional Brownian motion (which is also a fractal), basically Brownian motion which can allow for dependance on recent values or old values. Fractional Brownian motion has several applications in financial math, and it's been suggested that it replace geometric Brownian motion that's used in the famous Black-Scholes model. But, guess what? Fractional Brownian motion is a Gaussian process. Let's bash Mandelbrot for using something Gaussian. Before you rage, I suggest you learn what it is that you're raging against. | ||
|
corumjhaelen
France6884 Posts
Anyway, he had an appointment for in job in pricing. "So how do you price stuff ? -Well, basically we use Black Scholes -And does it work ? -Well it's pretty good because everybody uses Black Scholes" That doesn't tell the whole story, but hey at least they use funny mathematical concepts. We also had a heated argument because he believes that salaries are proportionnal or close to proportionnal to the added value you bring to your company. That was pretty funny. | ||
|
paralleluniverse
4065 Posts
On February 20 2013 20:56 corumjhaelen wrote: Just for fun, I have a friend studying finance (not that brilliant at maths and terrible at philosophy because he refuses to care, but a nice guy despite being in for the money). Anyway, he had an appointment for in job in pricing. "So how do you price stuff ? -Well, basically we use Black Scholes -And does it work ? -Well it's pretty good because everybody uses Black Scholes" That doesn't tell the whole story, but hey at least they use funny mathematical concepts. We also had a heated argument because he believes that salaries are proportionnal or close to proportionnal to the added value you bring to your company. That was pretty funny. That's what the most basic economic models say about wages. It's not true in the real world. | ||
|
corumjhaelen
France6884 Posts
On February 20 2013 20:59 paralleluniverse wrote: That's what the most basic economic models say about wages. It's not true in the real world. Hey he's the one who continued studying "economics" and complicated models, while I'm the one who's laughing at it. You know why he believes it ? Because then he can believe his pay is fair. | ||
|
JonnyBNoHo
United States6277 Posts
On February 20 2013 20:41 paralleluniverse wrote: I said that stress tests were one of the ways of dealing with tail risks. But there are others. Insurance companies deal with catastrophes all the time, (e.g. the Japanese earthquake and tsunami a couple of years ago). Why don't they go bust? No, AIG doesn't count, because it was the investment division, gambling with financial instruments, that blew up, not their insurance business. The main reason is because insurance companies know how to manage tail risk. Stress testing is one method. Another method is reinsurance, i.e. the insurance company buys insurance from another insurance company. The analogous concept in finance is using derivatives to hedge your position. Another is catastrophe bonds. The analogous concept in finance is a MBS. So why hasn't reinsurance and cat bonds destroyed insurance, the way that derivatives and MBSs have destroy finance? Because natural catastrophes, of the sort that insurance companies deal with aren't correlated, whereas everything is correlated in finance. A tsunami in Japan doesn't cause a hurricane in New York and a volcano eruption in Italy, whereas a large bankruptcy causes contagion to flow throughout the financial system. So this proves my point: it's not tail risk that was the main problem, it's correlation. It's possible to successfully deal with tail risk when things aren't correlated, just look at the insurance companies. Disasters aren't tied together but insured disasters can have the same contagion effect: Disasters lead to the insurance company selling assets to raise cash. Their asset sales drive down the value of the assets. As asset prices decline the insurance company loses its ability to make good on claims. Insured assets then become uninsured assets and their value declines. So on and so forth... Now in the real world this is never a meaningful problem, but I think that's because insurance just isn't that big relative to the overall market and not that they have some superior structure. | ||
|
sam!zdat
United States5559 Posts
On February 20 2013 20:32 paralleluniverse wrote: Financial math is not "flimsy mathematics". In fact, it's some of the most complicated and heavily research math out there. sounds like some high priesthood shit to me man. If your math is based on stupid premises (like the efficient markets hypothesis) it's flimsy, I don't care how complicated it is or how many fancy symbols it has. It's still bullshit. You say I rage against something I don't understand. Ok. Well. I'm an extremely intelligent person who doesn't want to devote his life to studying finance, but I've peen putting an entirely non-trivial amount of effort over the last year into understanding this stuff (because, you know, the total failure of this system has been a major event in my life), reading a lot of economic history and things like that, and I think it's a load of bunk, no matter how "complicated" it might be. Don't tell me it must be right and I must be wrong, because it's too "complicated" for a silly child like me to understand. That's just some obscurantist bullshit. You tell me why I should trust any financiers, when they get to fuck over an entire civilization and laugh to the bank. I don't have to understand all the technical details to know that you people are full of shit. Any idiot can see that the foundational premises of mainstream economic theory are philosophical non-starters. Rational choice theory? Efficient markets? please, tell me another In fact, the more "complicated" you tell me it is, the less I trust you, because I don't think any of you are smart enough to deal with something that "complicated" (you can't be, because you think people are rational actors, and any idiot knows that's not true). I think how "complicated" it is is precisely the problem, because you toy with something you do not understand, and the rest of us pay the price. The more "complicated" it is, the more chances it has for History to come in and burn your little tower to the ground. On February 20 2013 20:52 paralleluniverse wrote: Before you rage, I suggest you learn what it is that you're raging against. I AM, you smug suit, I just can't express it that well. Recommend me a book. When this first happened, and everybody was raging and occupying and stuff, I sat out. I said, "what do I know?" Then I learned some things, and NOW I'm raging. I got a right to be angry, because our world is FUCKED UP, and a big part of that is because of people who wear suits and have degrees in economics being fools and worse. edit: I'm well aware that I'm out of my territory here. I'm not pretending to really understand this stuff. But what I don't do is trust some suit when he waves his hands in my face and tells me he understands. Look Before 2008, some people said "we have some complicated math, we understand everything, trust us." Now they say, "we have some complicated math, we understand everything, trust us." What the fuck am I supposed to think? edit: I concede all questions about math, but I still think I'm right. | ||
|
corumjhaelen
France6884 Posts
We choose a model, compute a few derivative, find some extrema, compute a few formula and o wonder ! we can check that what we've found doesn't contradict our model's hypothesis, hurrah. We then vaguely inform the students that there are people out there who are trying to do number crunching and small scale experiments to test those hypothesis, which can usually be said as "could be worse". We then turn those students into traders. I wonder how that can go wrong. Edit : I forgot to add that wome ad hoc hypothesis were usually conjured to explain us that an obvious flaw in the model could be modelized a bit. I also that finance as models far more complicated than that, but I'm not convinced that using more complicated maths can even begin to close the gap between the models and reality. There were some interesting things described in my microeconomics course though, but they add more to do with psychology than pure economics anyway. | ||
|
sam!zdat
United States5559 Posts
| ||
|
aksfjh
United States4853 Posts
On February 21 2013 03:57 sam!zdat wrote: sounds like some high priesthood shit to me man. If your math is based on stupid premises (like the efficient markets hypothesis) it's flimsy, I don't care how complicated it is or how many fancy symbols it has. It's still bullshit. You say I rage against something I don't understand. Ok. Well. I'm an extremely intelligent person who doesn't want to devote his life to studying finance, but I've peen putting an entirely non-trivial amount of effort over the last year into understanding this stuff (because, you know, the total failure of this system has been a major event in my life), reading a lot of economic history and things like that, and I think it's a load of bunk, no matter how "complicated" it might be. Don't tell me it must be right and I must be wrong, because it's too "complicated" for a silly child like me to understand. That's just some obscurantist bullshit. You tell me why I should trust any financiers, when they get to fuck over an entire civilization and laugh to the bank. I don't have to understand all the technical details to know that you people are full of shit. Any idiot can see that the foundational premises of mainstream economic theory are philosophical non-starters. Rational choice theory? Efficient markets? please, tell me another In fact, the more "complicated" you tell me it is, the less I trust you, because I don't think any of you are smart enough to deal with something that "complicated" (you can't be, because you think people are rational actors, and any idiot knows that's not true). I think how "complicated" it is is precisely the problem, because you toy with something you do not understand, and the rest of us pay the price. The more "complicated" it is, the more chances it has for History to come in and burn your little tower to the ground. I AM, you smug suit, I just can't express it that well. Recommend me a book. When this first happened, and everybody was raging and occupying and stuff, I sat out. I said, "what do I know?" Then I learned some things, and NOW I'm raging. I got a right to be angry, because our world is FUCKED UP, and a big part of that is because of people who wear suits and have degrees in economics being fools and worse. edit: I'm well aware that I'm out of my territory here. I'm not pretending to really understand this stuff. But what I don't do is trust some suit when he waves his hands in my face and tells me he understands. Look Before 2008, some people said "we have some complicated math, we understand everything, trust us." Now they say, "we have some complicated math, we understand everything, trust us." What the fuck am I supposed to think? edit: I concede all questions about math, but I still think I'm right. You should probably tune into the people that had the "math" right in the first place, or have admitted they missed something back in 2007-2008. Also, those "simple" rules of economics you rail against are akin to many of the simple rules we have in physics, chemistry, and biology. You can approximate many of the physical laws, which are actually statistical probabilities, with very simple formulas and constants, and do so fairly well. You can present building a bridge as a few free-body diagrams to nearly anybody and have them understand the basics. However, building a bridge isn't that simple at all, and requires a TON of calculations of incredible complexity. The financial system works the same, and the sad part is that, in both areas, we've had trusted people screw up BIG and ruin thousands or millions of lives in both fields while KNOWING what was the right thing to do as a whole. | ||
|
corumjhaelen
France6884 Posts
In economics, there is no simple experiment that I know of, capable of testing small hypothesis. You have to directly test your model against a very complicated reality. And given that the bridge seems to fall down over over and over, I'm tempted to say that the hypothesis might suck, not the calculations. | ||
|
sam!zdat
United States5559 Posts
| ||
|
oneofthem
Cayman Islands24199 Posts
| ||
|
radiatoren
Denmark1907 Posts
On February 20 2013 20:41 paralleluniverse wrote: I said that stress tests were one of the ways of dealing with tail risks. But there are others. Insurance companies deal with catastrophes all the time, (e.g. the Japanese earthquake and tsunami a couple of years ago). Why don't they go bust? No, AIG doesn't count, because it was the investment division, gambling with financial instruments, that blew up, not their insurance business. The main reason is because insurance companies know how to manage tail risk. Stress testing is one method. Another method is reinsurance, i.e. the insurance company buys insurance from another insurance company. The analogous concept in finance is using derivatives to hedge your position. Another is catastrophe bonds. The analogous concept in finance is a MBS. So why hasn't reinsurance and cat bonds destroyed insurance, the way that derivatives and MBSs have destroy finance? Because natural catastrophes, of the sort that insurance companies deal with aren't correlated, whereas everything is correlated in finance. A tsunami in Japan doesn't cause a hurricane in New York and a volcano eruption in Italy, whereas a large bankruptcy causes contagion to flow throughout the financial system. So this proves my point: it's not tail risk that was the main problem, it's correlation. It's possible to successfully deal with tail risk when things aren't correlated, just look at the insurance companies. I am not 100% convinced that it was necessarily the financial worlds cross-correlations that caused the crash. I am more of a linchpin theorist, believing the higher order derivatives primarily and to a lesser extend the first order derivatives getting valued completely independently from the primary source of the stock. When the underlying industry is revealed to be worth far less than expected, it will have so much larger an extend on the market, when the ripple effects of "insurance" bonds go into effect. Also, having products with directly destructive value, as a bond basically counting on a crash, is probably not a good idea. The cross-correlation is among other things caused by the insurance culture among bonds and the total amount of money invested in the market. What I am saying is that the correlations you see on the stock markets is a direct result of the products on the market and how many different entities are affected. TBTF is a result of these correlations so why not set a max for how much a single cluster entity can buy of their own financial products without having physical assests to back it up and why not limit the insurance to certain levels of the physical background value of the stock? That would force them to invest broader and barring inversters fleeing it will be a much more stable market. | ||
| ||
