Ghost Patterns – D.E. Shaw’s Quant Strategy Explained

this hedge fund is a pioneer in quantitative investing d.e shaw is a firm that all quants are dying to get into managing 50 billion dollars dee shao is instrumental in creating the quant industry that now owns one third of all the u.s stock market for disha their composite fund which has been close to new investors is up more than ten percent that's beating the hedge fund average around eight percent but these days investors are happy if they see a double-digit return in the late 80s and early 90s many scientists and engineers have entered into the hedge fund industry they approach the market in a different way than economists economists believe that the market is efficient but mathematicians at ge shaw discover a lot of patterns that are not explained by economic logic there are statistical anomalies that are not random these are called ghost patterns they have made billions of dollars for quant firms like t.e shaw and renaissance technologies [Music] after earning his phd from stanford morgan stanley hired davi e shaw to create a computer system to support its quantitative trading morgan stanley at the time has a secretive analytical proprietary trading unit that ran a computerized effort to profit from short-run liquidity effects in stock markets the liquidity effect is a type of pattern that many quant firms tried to exploit in early 70s a hedge fund manager michael steinhardt discovered that a big sale order from a pension fund could push the stock price out of line provided that there were no information behind the sale economists always like to assume that all trading is between informed traders and liquidity providers so that means that if i have some information about a stock i will trade on it and then the market price will adjust according to this information i have and then the liquidity providers will also make a small amount of money for providing that transaction but michael steinhardt discovered that sometimes big transactions happen not because it was reacting to bad news but because the seller needed cash this is called the liquidity effect steinhardt could profit by buying and holding the stock until it rose back to its previous level morgan stanley's analytical proprietary trading unit aimed to find more trading signals like this with the advanced computational techniques but it became obvious to shaw that these patterns were often fading away too fast because they were easily caught on by others another example of these patterns is back in the 80s many commodity traders like to liquidate their positions before the market closed because they felt better holding on to cash at night the next day they would build up their portfolio back up again this behavior can be exploited you can buy their positions at night and then resell them the next day simple patterns like this are discovered because they have behavioral explanations shalt believed patterns that are difficult or impossible to explain are more valuable because others are less likely to discover and exploit them this approach is vastly different from morgan stanley who still operated like the academia we often see in academic finance they start out by having a lot of assumptions then they create a theory and then they look for data that justify that theory shell believed that this is not how real science is done the word theory means something different in science and it means something that's been very well tested quantum theory you know the the theory of gravity i mean you know gravity is there and the fact that it's a theory that means that we've tested these ideas over and over again finance people write down very complicated models and they're useful to think about but there isn't a correct one de shaw began investing in june 1989 having secured 28 million dollars in capital from a group of private investors the ghosts that shaw discovered were hard to explain when he found recurring patterns and printed them out there were no familiar terms that could be used to make sense of the anomaly a lot of people assume that quantz funds like dd shaw use high frequency trading but that's not the case they don't need to trade in milliseconds their advantage is these patterns they discovered instead of rely on speed pretty soon the profits started to roll in and shaw quickly expanded the firm to a much larger scale after finding profitable strategies and equities d e shaw discovered that there are more anomalies and derivatives in early 90s a group of economists started a hedge fund called long-term capital management company to exploit some inefficiencies in the derivatives market in a few years it has made around 40 percent annualized returns but failed catastrophically in the end they were valued at about 4.8 billion and this all took place in the weeks following russia's currency devaluation and bond default the federal reserve had to orchestrate a 3.6 billion bailout by the fund's 14 banks in order to calm fears that the firm's lenders and trading partners would be dragged down again these economists that manage this hedge fund made a fatal error that most of academic finance makes assuming a normal distribution simply put they underestimate the likelihood of rare events when it did happened it wiped out the entire fund and almost took down the financial sector with it shaw sought his chance his mathematicians were better at modeling than other market players without assuming normal distribution shaw's team came up with the options pricing model that gave him an edge in multiple markets by now shell had created a machine to discover anomalies in stock prices and derivatives some of strategies produce over 40 returns a year i think robert mercer who was the ceo of renaissance technologies summarizes the best when he says some signals that make no intuitive sense do indeed work the signals that we have been trading without interruption for 15 years make no sense otherwise someone else would have found them now the question is can some solo trader discover signals like this and profit from them professor marcos lupus de prato was managing a quad fund with about 10 billion typically what you want to do is you want to spend a lot of time identifying what are the features that are important for that strategy and once you come up with important features now you're going to build a strategy and at the last moment you're going to backtest you you want to spend most of your time in future importance a lot of people who do technical trading believe that they found some patterns but oftentimes they don't know much about statistical tests to validate those patterns they simply look at charge and try to you know make sense of these these things so i think what's important is to learn how to do statistical tests and for those of you who know mathematics and coding i think it's worth a try [Music] you

Channel: FINAiUS

Views: 304,211

Keywords: hedge fund, quant

Id: uK7dETASXsc

Length: 7min 41sec (461 seconds)

Published: Sat Dec 05 2020