Thread by @ReformedTrader

1/ Adaptive Markets (Andrew Lo)

"Finance isn’t physics, despite the similarities between the physics of heat conduction and the mathematics of derivative securities." (p. 10)

Feynman: “Imagine how much harder physics would be if electrons had feelings.”

amazon.com/Adaptive-Marke…

2/ "Academia, industry, and public policy have assumed rational behavior for so long that we’ve forgotten about the aspects that don’t fit as neatly into a mathematically precise framework.

"This oscillating between wisdom and madness isn’t a pathology. It’s human nature." (p.9)

3/ "The stock market punished Morton Thiokol, not on the day of the report (five months after the space shuttle accident), nor after Feynman’s brilliant live demonstration of the defective O-rings, but on January 28, 1986 itself, within minutes of the Challenger explosion.

4/ "The stock prices of Lockheed, Martin Marietta, and Rockwell International also fell, but their drops and overall volume traded were much smaller, and within statistical norms.

"Maloney and Mulherin were unable to find any evidence for insider trading on January 28, 1986.

5/ "Even more startling was the fact that the lasting decline in the market capitalization of Morton Thiokol on that day—about $200 million—was almost exactly equal to the damages, settlements, and lost future cash flows that Morton Thiokol incurred.

6/ "What took the Rogers Commission, with some of the finest minds on the planet, five months to establish, the stock market was able to do within a few hours." (p. 14)

7/ "With 28 students playing for a few hundred dollars of gift certificates for ten minutes, the correlation between the STOC rankings and traditional consumer-survey rankings was 0.85.

"The latter require several hundred participants, tens of thousands of dollars, and weeks.

8/ "We ran similar experiments with other consumer products—cars, laptop bags, and video game systems.

"In each case, we found that trading concepts provided very similar information to the much more expensive methods for measuring consumer preferences." (p. 41)

9/ "Our results showed that the variance of two-week returns was three times the variance of one-week returns, not twice the variance as predicted by the Random Walk Hypothesis.

"The odds of this happening by random chance alone were roughly 3 out of 100 trillion." (p. 48)

10/ "The first thing a junior trader learns is to “cut your losses and ride gains:” in the face of losses, fight the tendency to be too risk-seeking; in the face of gains, fight the tendency to be too risk averse. This is surprisingly hard to follow in the heat of the moment.

11/ "An initial loss causes an inexperienced trader to panic. Rather than owning up to the loss by liquidating the position, he takes the less psychologically painful route and increases his bet in the hope that markets will turn around and get him back to even." (p. 60)

12/ "Some economists claim regulatory forbearance (the tacit or active cooperation of regulators in overvaluing bank assets to avoid violating minimum capital requirements) is partly responsible for the recent financial crisis, offering elaborate explanations for why it occurs.

13/ "But a more mundane possibility, one we shouldn't dismiss, is loss aversion: a sure loss to the regulator if she calculates that a bank’s assets have declined, and a riskier but less psychologically painful alternative if she maintains the older, higher estimate." (p. 62)

14/ "Why do people find order within random patterns? They unthinkingly take a small sample as representative of the whole, using the representativeness heuristic." (p. 67)

15/ "The tendency to try to predict everything is both a blessing and a curse. It explains why we’re able to survive in environments ranging from the Arctic to the surface of the Moon, but it also means we sometimes incorrectly attribute meaning to unpredictable events." (p. 69)

16/ "The skepticism we encountered was emblematic of the state of academic finance. We spent ten years trying to explain away our results as a quirk of the data.

"Despite our best efforts, we were unable to explain away the evidence against the Random Walk Hypothesis." (p. 71)

17/ "Price swings in stock XYZ this week had predictive power for forecasting the swings of stock ABC next week.

"The academic community had ignored earlier studies rejecting the random walk. They were never included on the reading lists of any of our graduate school classes.

18/ "Our colleagues, like ourselves, had been trained to study the data through the lens of classical market efficiency. All of us in the academic community lived in a collective fog.

19/ "I reluctantly concluded that the conflict between the broad, almost universal support for the Random Walk Hypothesis and our empirical findings was largely due to the near-religious devotion of economists to the Efficient Markets Hypothesis.

20/ "Not only had the Efficient Markets Hypothesis become an article of faith for many economists, it had hardened into a tenet of dogma.

"Any departure from this paradigm was deemed heresy and unceremoniously dismissed." (p. 73)

21/ Related reading:

In Defense of Troublemakers: The Power of Dissent in Life and Business (Charlan Jeanne Nemeth)

https://twitter.com/ReformedTrader/status/1368396960688869381

Science Fictions (Stuart Ritchie)

https://twitter.com/ReformedTrader/status/1336398442403307521

The Structure of Scientific Revolutions (Thomas Kuhn)

https://twitter.com/ReformedTrader/status/1331998734276653059

22/ "Orthodox assumptions capture human behavior well enough that most economists instinctively reach for those explanations. Few economists really believe that individuals actually behave rationally all the time, but all economists are trained in methods that assume they do.

23/ "Not every outsider with these new ideas participated in the intellectual debate. Instead, some took advantage of the 'huge profit opportunities in the U.S. equity market' " (the ones that EMH suggested should not be there). (p. 74)

24/ "Slovic found a bias: If the potential risks and benefits of a policy are framed to provoke a negative (positive) emotional response, people overweigh the risks (benefits) and downplay the benefits (risks). Our fears cause us to exaggerate risks that viscerally affect us.

25/ "We think our risk of dying in an accident is twenty-five times greater than dying from a stroke; in fact, we’re twice as likely to die from a stroke as we are in an accident. As a result, we over-focus on emergency medical treatment and neglect healthy eating and exercise.

26/ "Slovic found that not even expert policymakers are immune to this bias, which he calls the affect heuristic." (p. 84)

27/ "The modern slot machine is designed to elicit specific neurological responses.

"The player is given the illusion of control. The payoffs are fast, in order to keep a tight connection between a win and the rush of dopamine. Losses are deliberately displayed as near-misses.

28/ "The brain still triggers reward circuitry in the event of a near-miss, just not as intensely as for a win. The environment around the slot machines is designed to minimize anxiety and other negative emotions. These factors combine to make slot machines addictive." (p. 91)

29/ "Traders who described more intense reactions to both losing and making money performed worse. Those who scored higher on “internality”—the tendency to ascribe the causes of various events in their lives to their own doing versus random chance—also performed worse.

30/ "These patterns tell us something about the stuff good traders are made of: more controlled emotional responses, including the ability to refrain from blaming (or lauding) oneself too much for trading outcomes." (p. 95)

31/ "Without certain emotional abilities, the ability to process risk is impaired. Players with this type of brain damage see high reward but fail to account for the even higher risk.

"When the ability to experience emotions is removed, behavior becomes less rational." (p. 108)

32/ "It’s not very difficult to construct a scenario where the correct knowledge regarding another individual’s intentions five levels removed has financial implications (complicated M&A deals, exotic financial derivatives, the picks and trades of the NFL draft).

33/ "But if it’s impossible for all but a few chess grandmasters to hold such a chain as a single thought—in the same way that a 3-year-old child can’t understand that his mother doesn’t know where his blanket is—how can investors always act rationally? Short answer: they can’t.

34/ "The history of markets is filled with investors going wrong with utter confidence in the soundness of their judgment—until brought down by information just beyond their range of understanding. Our rationality is too limited for EMH to hold in every context." (p. 111)

35/ "As Tversky and Kahneman demonstrated, we generally don’t use logic or mathematics to make these snap judgments: we use heuristics instead. But when we’re asked to explain ourselves, we can usually construct a rational-sounding reason for our behavior." (p. 113)

36/ "Gazzaniga provides numerous examples where a split-brain patient is stimulated in some manner, and when asked to explain his reactions, the patient creates a narrative, one that seems coherent but is in fact a wildly irrelevant and incorrect explanation.

37/ "The left hemisphere constructs a narrative that fits the observed data—but it doesn’t always do so correctly.

"This ability to construct a narrative is central to what we mean by “intelligence.” " (p. 117)

38/ "Comparing pairs of Swedish fraternal and identical twins, Barnea, Cronqvist, and Siegel found that 1/3 of the observed investment behavior could be attributed to genetics: 29% of stock market participation, 32% of share of equities, and 38% of portfolio volatility." (p. 161)

39/ "To a nonacademic, back-and-forth discussion between competing narratives may seem incredibly tiresome and frustrating—we just want the answer, please! But this misses the point—it’s the ongoing exchange of ideas that ultimately leads us to the answer." (p. 167)

40/ "Simon assumed that every time an individual made an economic calculation toward a decision, it exacted a cost on the individual, which could be expressed monetarily. (Think about the wear and tear it takes to do our taxes; we’re willing to pay someone to do them for us.)

41/ "We calculate toward the best solution until a breakeven point where additional benefits are balanced by the cost of getting there.

"Individuals don't optimize—they satisfice, making heuristic decisions that are 'good enough.' Simon called this bounded rationality." (p. 179)

42/ "Simon’s critics dominated discussions in economics about satisficing for decades. Satisficing was rarely mentioned, and when it was, it was brought up as yet another failed theory against the reigning orthodoxy of the Efficient Markets Hypothesis." (p. 182)

43/ "Emotion is the primary feedback mechanism that causes us to update our heuristics. Love, hate, sympathy, jealousy, anger, anxiety, joy, grief, and embarrassment all serve useful purposes in telling us something about our environment and how we may wish to alter our behavior.

44/ "My heuristic has evolved thanks to social feedback. Wearing a suit and tie to teach my MBA classes is good form; wearing a suit and tie to a research meeting with academic colleagues is pretentious.

"Our environment and life history actively shape our behavior." (p. 184)

45/ "Under the Adaptive Markets Hypothesis, individuals never know for sure whether their current heuristic is “good enough.” They make choices based on their past experience and their “best guess” and learn by receiving positive or negative reinforcement from the outcomes.

46/ "This can explain behavior that looks irrational. If the environment changes, the heuristics of the old environment might not be suited to the new one. If individuals receive no reinforcement from their environment, positive or negative, they won’t learn.

47/ "If they receive inappropriate reinforcement from their environment, individuals will learn decidedly suboptimal behavior. And if the environment is constantly shifting, it’s entirely possible that individuals will never reach an optimal heuristic.

48/ "But the Adaptive Markets Hypothesis refuses to label such behaviors “irrational.” It recognizes that suboptimal behavior is going to happen when we take heuristics out of the environmental context in which they emerged." (p. 188)

49/ "An investor may buy near the top of a bubble because she first developed her portfolio management skills during a bull market.

"Behavioral biases are heuristics we’ve adapted from nonfinancial contexts that we misapply when we use them in financial settings." (p. 189)

50/ "In the 1960s, automobile safety became an important social issue in the U.S. We passed laws mandating that cars have seat belts, padded dashboards, safer steering columns, and safer windshields to cut down on the number of deaths caused by fatal head and chest injuries.

51/ "But when UChicago economist Sam Peltzman looked at the data in 1975, he concluded that any increases in safety were offset by a worsening of driver behavior. The increase in pedestrian deaths offset the decrease in driver deaths that followed the new safety features.

52/ "A number of researchers disagreed with Peltzman’s conclusions, arguing that his study didn’t take into account a number of confounding factors. Since it’s very difficult to run controlled experiments for car-accident fatalities, this debate has continued for decades.

53/ "In 2007, Russell Sobel and Todd Nesbit cleverly identified a venue where all the cars and drivers experience virtually identical conditions, so that the only thing that matters to the drivers is getting to the destination a little bit faster: NASCAR races.

54/ "In this context, every time a new safety device was introduced, the number of accidents actually increased. Drivers adapted to the new safety measures by adjusting the risk preferences of their behavior accordingly." (p. 205)

Thread:

https://twitter.com/ReformedTrader/status/1257849272231317504

55/ "Physicists can explain 99% of all observable physical phenomena using Newton’s laws of motion. Economists wish we had three laws capable of explaining 99% of all observable behavior in our purview. Instead, we have ninety-nine laws that explain 3% of all economic behavior.

56/ "So we sometimes cloak our ideas in the trappings of physics. We make axioms from which we derive seemingly mathematically rigorous universal economic principles, carefully calibrated simulations, and the very occasional empirical test of those theories.

57/ "Several physicists have pointed out that if economists genuinely envied them, they’d place greater emphasis on empirical verification of theoretical predictions and show less attachment to theories rejected by the data, neither of which characterizes our profession." (p.209)

58/ "But many questions in economics did become more manageable. We can read the classics of economists who came before Paul Samuelson—great thinkers like Adam Smith, John Stuart Mill, Karl Marx, or John Maynard Keynes—and become lost in the abstractions of their lengthy prose.

59/ "The intellectual environment of economics was ripe with problems that could be solved with these ultra-mathematical techniques. What’s more, this borrowing from physics was also financially profitable." (p. 211)

60/ "Samuelson's protégé, Robert C. Merton, created much of what is now known as financial engineering as well as the analytical foundations of multi-trillion-dollar industries: exchange-traded options, OTC derivatives and structured products, and credit derivatives." (p. 211)

61/ "The mathematization of economics is now largely complete. However, even as Samuelson was sowing the seeds of physics envy for future generations of economists, he was also keenly aware of the limitations of mathematical deduction in economics." (p. 212)

62/ "Simon’s bounded rationality clearly beat Samuelson’s neoclassical theory of utility maximization in psychological plausibility, but after Samuelson, most economists simply weren’t interested in realistic representations of internal states.

63/ "They wanted a theory of economics as powerful and abstract as the nuclear physics that had given the United States the atomic bomb. They distrusted the measurement of the subjective, and they distrusted psychology as a whole." (p. 213)

64/ "Economists applied their highly mathematical theories of rational behavior in every conceivable way, not realizing that their environment was becoming depleted of problems where those theories were appropriate." (p. 213)

65/ "Within our new framework, market behavior adapts to its financial environment. An efficient market is the steady-state limit in an unchanging environment. Such an ideal is unlikely to actually exist, but its performance can be approximated under certain conditions." (p. 221)

66/ "Buyers and sellers don’t use all available information to make their decisions—they use *some* information and heuristics instead. These heuristics can be extraordinarily sophisticated, like the quantitative strategies of Shaw and Simons, but they’re still heuristics.

67/ "Today’s financial markets are still distant from a theoretical end-state of perfectly efficient markets. Investors as different as Warren Buffett and Jim Simons consistently out-earn the index funds favored by EMH despite using very different investment strategies." (p. 225)

68/ "Buffett confronted EMH in 1984. If fund success were the statistical equivalent of orangutans flipping a coin twenty heads in a row, they should be distributed evenly in population. But if they “came from one zoo in Omaha, you would be pretty sure you were on to something.”

69/ "If most hedge fund successes used the same underlying investment principles, you could be pretty sure it’s no accident. But Buffett failed to convince his debating opponent, Michael Jensen, or academia more broadly." (p. 234)

Related reading:

https://twitter.com/ReformedTrader/status/1217884016797310976

70/ "To hedge against the risk of handling large block trades, Morgan’s block traders would sell short a smaller chunk of a related stock in the same industry.

"Traders were able to profit from the brief price spread differences caused by the block trades." (p. 235)

71/ "Shaw looked for brilliant people with strong backgrounds in the mathematical or physical sciences rather than finance professionals.

"The insight that a hedge fund might be modeled after an academic research organization was a key contribution to his success.

72/ Shaw: “The obvious mathematical techniques that someone would probably try first looked like they’d been arbitraged out of the market long before. But there was still a fair amount of juice out there to be squeezed.”

73/ “We were able to run randomized controlled trials in which we could compare two models or parameter values to see which one performed better in actual trading. Analyzing the results of live trading taught us things that couldn’t be learned by studying historical data.“

74/ “As we discovered new anomalies, we also benefited from a second-order effect. If the profit from a given single effect was exceeded by the transaction cost that would be incurred to exploit it, it would be a mistake for anybody to bet on that effect in isolation.

75/ “Once we’d identified a number of small inefficiencies, though, the aggregate profit could break through the transaction cost threshold. This allowed us to extract profits from inefficiencies that were too small for most traders, creating a barrier to entry for competitors.”

76/ “Effects tended to disappear over time. Anomalies that had previously generated significant profits stopped making money, and you had to discover other, more complex effects that people hadn’t found. Quantitative trading became more challenging each year.” (p. 239)

77/ "By Sept 25, 1998, LTCM’s balance-sheet leverage ratio had risen to 250-to-1 and was rapidly shooting to infinity. Off-balance-sheet items added to its troubles.

"Its positions represented well over 10% of some futures markets." (p.243)

More on this:

https://twitter.com/ReformedTrader/status/1271108014498385920

78/ "LTCM’s collapse caused most hedge funds to rethink their risk models. Some reevaluated their entire approach to investing.

"The strategic adaptability of the hedge fund has made it resilient in the face of unforeseen financial events." (p. 243)

79/ "At first, high-frequency traders made windfall profits, since human specialists were sluggish in comparison. However, there ultimately came a point when HFTs were mainly competing with each other. To succeed, they had to invest in faster and more expensive hardware.

80/ "HFT became a mature industry, with low margins on trades and low overall profits. It had also become highly adapted and extremely sensitive to changes in the regulatory environment, for instance, to a Tobin or transactions tax, and even to the speed of light." (p. 245)

81/ "Markets aren’t guaranteed to become more efficient. The drive for returns sowed the seeds for an even bigger crisis than LTCM had been. The financial environment of 2007 was quite different than that of 1998; the Quant Meltdown was a warning sign of a regime shift." (p. 293)

82/ "Near the peak of the housing market in 2006, a Russian émigré remarked, “This place seems very rich, but I never see anyone making money.” Krugman’s response: “These days, we make a living selling each other houses paid for with money borrowed from the Chinese.” " (p. 298)