> The efficient-markets hypothesis says markets are rational and self-regulating, but it doesn’t account for crashes and crises
It's also a computationally insane assertion, if it were true then markets are capable of computing NP in P (which even with the human element would still be an impressive find; especially considering the `n` it's commonly proposed over in "formal" economic settings is in the area of tens of thousands). It's like people who claim they made a perpetual energy machine are claiming things that are physically insane. Except it's not just some cranks, it's entire fields that haven't caught up to the ideas computer science has discovered.
The adaptive market hypothesis appears to resolve this by tying the efficient market hypothesis to the biology of the computational unit. Since biological systems (appear to) obey the principles of computation, the issue is fixed. Overall it seems a better way forward by connecting pseduo-sciencey intuitive aspects of economics (that violate what we now know about the properties of information) to real science.
I've seen this paper enough to warrant a comment. The paper makes no reduction that the market is computing in P, so it's unclear to me how this conclusion of P=NP iff EMH is warranted by the paper.
Perhaps more clearly showing the connection between 'complexity' in the general sense and EMH is this paper[0], showing that even (eps, eps)-weak (or approximate) Nash equilibria are communication-hard (exponential in the number of players), would be the first that has convinced me that markets are highly unlikely to be efficient.
> It's also a computationally insane assertion, if it were true then markets are capable of computing NP in P
Unless I missed it, they don't assert that markets are optimal, which seems to be what you're saying, but merely efficient and thus, it would be hard (but possible) to do better.
> Financially, the “economic calculation problem” of von Mises (1920) and Hayek (1935) suggests,
among other things, that, even if a free market is not perfectly efficient, it will certainly be more
efficient than a regulatory or government alternative. In other words, even if mispricings
occasionally occur, most of the time they are smaller than any other alternative system
So the paper possibly disproves a particular kind of efficiency (weak form efficiency), but that's not the colloquial meaning to which I was referring.
After doing my MSc in Political Economics and studying lots of Game Theory, I became more interested in computer science. I remember watching a Stanford lecture online on computational game theory. He explained many games are NP hard, and while this is only my own experience, we never once covered that idea in any of our courses on political/economic game theory. To be honest, I'm not sure any of the older economists or Political Scientists were all that familiar with it (I'm sure the incredibly bright younger post-docs were though).
That really blew my mind, because up until that point I guess I had sort of taken it on faith -- without really getting it -- that these game models were accurately representing reality. I guess it sounds obvious after the fact that they aren't, but it was one of the biggest insights I've ever experienced.
Also, the unacknowledged inconsistency of his arguments is ridiculous: ~"Unpredictable events happen more frequently than people expect. I'm 100% right about this prediction of the future and you are an idiot if you disagree".
I just finished all 3 of Taleb's major books, and I took a slightly different interpretation of his Black Swan concept. I am very interested in reading criticisms of his that focus on the material -- most that I've read thus far discuss his writing style, or lack of rigor. (The second is certainly a valid criticism, but I'm looking to engage with his ideas, not the man himself or the broad scientific validity.)
Taleb's Black Swan concept is that unpredictable events happen more frequently than people expect, and have an outsized impact on the outcomes of a model. Events that are not predictable are not included in the predictive model. Take the frequency of event and exclusion from the model, add in asymmetric results (small changes to input parameters can lead to huge changes in outcome), and that's his theory.
- it's hard (impossible) to predict population statistics from a sample with size 1, ceteris paribus. Over the short term, you have sample size 1 for rare events, over the long term ceteris paribus doesn't hold.
- people "coerce" everything into a normal distribution in order to get a handle on things which is of course wrong if the thing you look at is not normally distributed. Finance comes to mind.
Sorry the two sentences in the quasi-quote are separate things. The first sentence is his book 'Black Swans' and the second is his Twitter feed. I should have made that more clear.
And additionally, Taleb is enormously pompous. His books have some interesting ideas in them, but they’re enveloped in a cloud of his farts. Makes him very tedious to read.
I like my authors to be pompous. Because anybody whose not will not write a 2000 page book. You have to have an internal belief that people want read your ramblings.
Has anyone read Adaptive Markets? If so would you recommend it? I'm concerned about this line from the article:
> Lo’s next step is to develop mathematical models to test the ability of his hypothesis to explain market behavior.
Is there any experimental data to back up his theoretical framework? It's a very tempting idea to buy into (people in general work things out, on an individual level they make a lot of mistakes). I worry about his arguments following logically but falling apart once his theories have been tested.
I found Adaptive Markets to be pretty weak. I was excited to read it, but I didn't finish it, because I found the content to be relatively facile and substanceless. YMMV, though.
If he is the Lo I think he is then his textbook on econometrics (Mckinsey & Lo?) is one of the best. I would guess econometric studies support his insights already but he tries to formulate experimental results in a consistent model that aligns with his theory.
Since undergrad while taking economics classes, I thought this was entirely obvious. And honestly, I still do. To assume everyone is a rational actor is to assume everyone is perfect: they don't make mistakes, they don't have asymmetric information, etc.
You can already see the changes that this articles suggests should happen in practice with high speed trading. I'd love a more detailed article, because this describes something obvious and vague.
You may be misinterpreting things. Relevant from Wikipedia: "Lo argues that much of what behaviorists cite as counterexamples to economic rationality—loss aversion, overconfidence, overreaction, and other behavioral biases—are, in fact, consistent with an evolutionary model of individuals adapting to a changing environment using simple heuristics."
And as an aside, I am pretty tired of people saying things like this, without fail, in every thread about economics. A simplified model is better than no model at all. People have been trying to account for irrationality for ages. It's not particularly astute to recognize that individuals are not perfectly rational.
> A simplified model is better than no model at all.
This is true in physics where the system being modelled is governed by stable laws. You can confirm the model's predictive usefulness by experiment. If you test the model and find its limitations, you expect them to hold in 50 years' time.
Economic models, however, are themselves part of the economy. The ideas they are built on and their predictions become part of wider knowledge and influence the very thing the model is trying to predict or explain, possibly reinforcing or weakening certain behaviours. If a simplified model becomes widespread it can lead to blind spots in our understanding which cause systemic issues.
Of course that doesn't mean the model was bad or unsuitable for its original intent. But a simplified model can, in the wrong hands, be a net negative.
That is an interesting point, that people have bought too much into this simplified model s.t. reality bends towards the model. I don't know if I blame the model for that - like you say in your last sentence.
But what's the alternative besides starting with simplifications and iterating?
> Economic models, however, are themselves part of the economy. The ideas they are built on and their predictions become part of wider knowledge and influence the very thing the model is trying to predict or explain, possibly reinforcing or weakening certain behaviours.
Which is the very problem "rational expectations" modeling [0] is trying to avoid. Rational expectations models don't 'know' anything (systematic) that the actors in it don't.
The problem is not that people are irrational, the problem is that the "rationality" is a convenient mathematical ideation, not a normative statement on how people ought to behave. It's just a terrible term.
"Rational" is definitely used badly. It's originally about ratio which means simply that we can compare things with mathematical accuracy, the assumption being that we use such processes in all sorts of thinking, but that's absurd. That it has come to mean something like "reasonable" is unfortunate.
Good luck eliciting those numerical assignments, and then tracking them as they change in continuous time based on what your rational agent ate for lunch, how his sex life has been recently, et al.
A simple model is better than no model when it makes useful predictions over a well defined domain.
The main problem with most models of market economics is that they only work when nothing interesting is happening.
Probably the best, as in most practically useful, idea from market economics has been that of arbitrage freedom. And even that falls apart in low liquidity scenarios which often characterise severe events.
"I am pretty tired of people saying things like this, without fail, in every thread about economics. A simplified model is better than no model at all."
Funny. I'm pretty tired of people saying that a terrible model that makes terrible predictions is better than no model.
The unifying feature of perfectly rational, perfect information and a perfectly competitive market is not that they are the best way to simplify economic models, it's that they're the best ways of creating models that conceal true sources of profit.
nor is it particularly astute to create an entire discipline used to justify our entire social order on a theory that is known at the onset to be incorrect, all the while claiming it is more 'rational' and less 'superstitious' than the ones that came before it..
> nor is it particularly astute to create an entire discipline used to justify our entire social order on a theory that is known at the onset to be incorrect,
This is precisely what makes your line of argument tiresome: you fail to understand the very basics of what models are and how they are used.
What determines if a model is valid or not is its error margin for the intended use. Even if a model fails to draw accurate predictions, its ability to express cause-effect consequences is enough to demonstrate the model's usefulness. It makes absolutely no sense at all, and flies in the face of reason, to criticize a model because it might not be very accurate at representing a corner case. That's a very ignorant thing to claim.
To paraphrase George Box: "All models are wrong but some are useful". It seems you failed to understand the latter.
> This is precisely what makes your line of argument tiresome: you fail to understand the very basics of what models are and how they are used.
no, this is precisely what makes your line of argument argument tiresome - choosing one inherently biased and flawed model over another because you prefer the way that the biases look..
and actually I wasn't "arguing" for/against anything - but pointing out logical inconsistencies in an already occurring exchange..
Which theory specifically is used to "justify our entire social order"? How do we know it's incorrect, and what are the criteria for a successful theory?
Well, I don't intend to defend the blanket phrasing of GP, let's take for example GDP and it being taken seriously as a measure of a country's progress.
I would wager just about any undergrad could come up with a superior measure to GDP.
rational free market theory; humanistic positivism, etc.
in the case of rational free market theory, even OP pointed out the logical problems inherent in this theory, i shan't repeat here since it doesn't deviate from the thread
Rationality and adaptive markets hypothesis are not mutually exclusive. Moreover you seem to be attacking a strawman of what economists mean when they use the term, "rational actor". Economists do not actually assume everything is perfect or that the market is perfectly efficient - not even Fama postulates that. Rational behavior is an approximation, and economists are such approximations do not perfectly model the market. That doesn't mean they don't have any predictive ability or empiricism.
I also didn't see any changes suggested by the article - there doesn't seem to be anything in it that's at all prescriptive; how are you drawing a connection to HFT?
The rational actor model implies things like transitivity of preferences. The rational actor model, even an "approximate" model which accounts for errors and limited information, is fundamentally broken because we know that humans _systematically_ violate such basic premises as transitivity.
Put another way, we could just as easily say that non-human animals approximate rational actors because populations rather quickly approach efficient ecological equilibriums. But that sort of framing is not very helpful.
Absolutely not. I keep seeing this simplification thrown around and it's patently false.
The uniform spherical cow analogy works better when saying "consider a nation-state as a single actor in a global economy."
The rational actor assumption is more along the lines of the continuum approximation in fluid dynamics: that even though we don't have the power to model each individual molecule, we can do well enough by applying heuristics on top of that (pressure, density, etc.) that we can reason out some cause-effect behavior by measuring and tracking those higher-level quantities. We can even derive equations from them, though they may not have analytical solutions.
The problem with this is of course not just the inhomogeneity of the environment with respect to these higher-order quantities, but with the molecules themselves. This combined with strong nonlocal effects makes economics a really really difficult science to pin down mathematically, so all we seem to be able to do so far is fit the data to one of a few common models and wipe our hands clean of it.
No one assumes or thinks this. Economists know about mental illness for starters.
They assume there exists enough rational actors within the market such that it's run by rational actors.
What you'd need to think is.. no one in the market is a rational actor or a lessor form there are not enough ration actors in a market to keep it running, to think differently.
What is the legality of making something to help people not pay for stuff? Not trying to be confrontational, I sincerely don’t know + don’t feel sorry for them.
I haven't tried it either but I'm pretty sure there is no legality issues outside of the US, probably dubious legality inside the States but I wouldn't want to go up against their lawyers to prove it..
1. Economic theories can be empirically descriptive without being constructively prescriptive (much like other theoretical disciplines, such as math or physics),
2. You're reading about his work from what is effectively a watered down blurb on a Bloomberg webpage,
3. He actually is the founder and chief investment officer of a fund with over $5B AUM, so yes, he likely is "rich".
It's also a computationally insane assertion, if it were true then markets are capable of computing NP in P (which even with the human element would still be an impressive find; especially considering the `n` it's commonly proposed over in "formal" economic settings is in the area of tens of thousands). It's like people who claim they made a perpetual energy machine are claiming things that are physically insane. Except it's not just some cranks, it's entire fields that haven't caught up to the ideas computer science has discovered.
The adaptive market hypothesis appears to resolve this by tying the efficient market hypothesis to the biology of the computational unit. Since biological systems (appear to) obey the principles of computation, the issue is fixed. Overall it seems a better way forward by connecting pseduo-sciencey intuitive aspects of economics (that violate what we now know about the properties of information) to real science.