Re: MI5 - Missão: 007 / Sistema de Trading
Enviado: 25/4/2020 14:41
Caro Xacal:
Sim, é verdade que o Medallion Fund foi fechado à subscrição do público ao fim de 11 anos em 1993, na altura a gerir perto de 10 biliões USD. Os retornos eram (e continuam a ser!) excepcionais e os pedidos de adesão eram muito elevados, provavelmente foi o motivo para ser fechado aos clientes externos mas outra razão pode estar relacionada com o facto de que é cada vez mais difícil garantir retornos de excelência à medida que o valor dos investimentos a gerir sobem para patamares muito mais elevados. Basta pensar: porque motivo o fundo principal de 200 biliões USD da Berkshire Hathaway do Warren Buffett nunca mais conseguiu bater o mercado? Porque o fundo com aquela dimensão é quase o próprio mercado!
De qualquer maneira os fees no Medallion para os amigos e colaboradores são de 5% "front management fee" do total investido na data de subscrição e de 44% "performance return fee" sobre os lucros obtidos no resgate, cerca de metade fica na casa gestora e quem resgata com o resto fica milionário!
Por curiosidade existe uma entrevista muito interessante no MIT com o Jim Simons, com duração de mais de uma hora, que pode ser vista aqui abaixo para quem tenha tempo livre disponível, de certeza que vai dar por bem empregue o tempo gasto:
https://www.youtube.com/watch?v=srbQzrtfEvY
Ao pesquisar sobre a performance actual dos fundos da Ren Tech do Jim Simons neste ano de 2020 encontrei este curioso artigo desta semana no Institutional Investor:
The Famed Medallion Fund
Is Crushing It. Other
RenTech Funds, Not So Much.
The disparity — a performance difference of 17 to 19 percentage points — is “really surprising,” according to Renaissance Technologies skeptic Bradford Cornell, a professor emeritus at the University of California Los Angeles.
Amy Whyte
April 21, 2020
Not even a crisis can bring down Renaissance Technologies’ market-leading Medallion fund. However, the same cannot be said for the quantitative firm’s other hedge funds — the ones meant largely for outside investors.
According to the Wall Street Journal, the famous — and famously secretive — Medallion fund was up a stunning 24 percent year-to-date through April 14. This included a 9.9 percent gain in the month of March, when stock prices plunged.
Yet Renaissance Technologies’ other funds have not fared so well in 2020. Even after bouncing back in the first half of April, the three funds — Renaissance Institutional Equities Fund, Renaissance Institutional Diversified Alpha, and Renaissance Institutional Diversified Global Equities Funds — were each down between 7 and 9 percent for the year through April 17.
The disparity — a performance difference of some 17 to 19 percentage points — is “really surprising,” according to Bradford Cornell, a professor emeritus at the University of California Los Angeles.
The professor’s surprise is, perhaps, not surprising at all.
Cornell previously analyzed the performance of the Renaissance Technologies funds in a brief paper entitled “Medallion Fund: The Ultimate Counterexample?” In the paper, he wrote that the performance of the Medallion fund — which between 1988 and 2018 delivered a gross annualized return of 66 percent — was “extraordinary.”
Cornell writes in the paper’s abstract: “The performance of Renaissance Technologies’ Medallion fund provides the ultimate counterexample to the hypothesis of market efficiency. To date, there is no adequate rational market explanation for this performance.”
Turning to the returns of the REIF and RIDA funds, however, Cornell found that they were “relatively mundane and in no way comparable to Medallion.”
Wall Street Journal reporter Gregory Zuckerman, the author of The Man Who Solved the Market, says the large gap in returns can be explained by the differences in strategies among the four funds.
“Discrepancies between Medallion's returns and those of other Renaissance funds catch the eye, but I've never seen evidence something nefarious is going on — neither has the SEC, which spent two years in their offices after the Madoff scandal,” he said in an emailed statement. “Medallion has short-term holding periods, RIEF and the other funds search for longer-term aberrations and own smaller stocks and other investments — things Medallion shies away from.”
According to fund documents, the Medallion fund employs a short-term, quantitative trading strategy across multiple asset classes. These include global equities, futures, commodities, and currencies, according to a person familiar with the fund. It also tends to have high turnover and significant leverage.
The Renaissance Institutional Equities Fund, by comparison, only trades in equities, and holds stocks for long periods of time, according to fund’s registration document.
As Institutional Investor previously reported, the REIF fund was created to generate gross annual returns of 400 to 600 basis points — or 4 to 6 percentage points — above the S&P 500 over rolling three- to five-year periods.
The fund was up 8 percent for the month as of April 17. Year-to-date, it was down 7 percent — roughly 400 basis points better than the S&P 500, which was down by over 11 percent for the year through April 17.
The Renaissance Institutional Diversified Alpha fund, meanwhile, trades equities, derivatives, and various instruments in the global futures and forwards markets, according to fund documents. Like REIF, the RIDA fund holds significant individual positions, usually for long periods of time.
It was up 1.55 percent for the month through April 17, for a year-to-date loss of 9 percent.
The Renaissance Institutional Diversified Global Equities Funds, which trades equities and derivatives, was also in the red as of April 17, despite a 3 percent gain for the month thus far. According to fund documents, the RIDGE fund seeks to be market neutral by maintaining low levels of beta, or exposure to the broader market.
But as the firm admitted in the fund registration document, “the beta models in recent volatile markets have not performed as expected.”
Overall, the returns are in stark contrast to that of the Medallion fund — a fund that has been closed to external capital since 1993.
“If they have such secret sauce, how could the public funds be down so much?” Professor Cornell said in an email. “I wish I had an answer.”
Explosively, professor Bradford Cornell’s recent brief analysis of the performance of Renaissance Technologies’ Medallion fund states that “to date, there is no adequate rational market explanation for this performance.”
I find this conclusion interesting — not because of its contribution to our understanding of the Medallion fund’s performance (such efforts have devolved into a parlor game), but because it accidentally reveals an implicit worldview that, upon reflection, is a root cause of the dismal record of active management.
This belief rests on the conceit that investing is essentially a human activity — and therefore human reason must be able to explain the results of such activity.
There is a continuum of expressions of this worldview, ranging from the dogmatic assertion that active investing must be based on either cash flow analysis or short-term price movement to authoritarian claims that there can be no investment activity without human talent.
Through dictatorial educational programs and a homogeneous workforce, we have reached a point where this perspective is unquestionably accepted. This acceptance allows us to pretend that everything worth knowing is already known and in use, making any alternative unimaginable.
In investing, we, too, have our own “scripture” that grounds our perspective, and, though not revelatory, it has been unshaken for the past 60 years.
For example, since Michael Jensen’s 1968 landmark study of the performance of 115 mutual fund managers, we have held the view that active management cannot consistently provide returns better than a relevant passive proxy.
Yet the “extraordinary” and “unprecedented” performance of the Medallion fund presents us with a phenomenon that contradicts this fundamental belief. “In forty plus years of reading hundreds of papers on investment anomalies, including some that benefited from data snooping and ex-post selection bias, I have never seen any performance approaching that reported by Medallion,” professor Cornell concluded in his paper.
Ultimately, Cornell wrote that he could come up with no “convincing” explanation for the Medallion fund’s outsize returns, noting that even if Medallion was simply better at trading than any other fund, “the returns are so large, it stretches that explanation to the limit.”
Cornell’s incredulity is not an unusual reaction to the Medallion fund, which has long baffled industry insiders and observers. The secret to Renaissance Technologies’ performance has been debated in news articles and academic circles and on online message boards — and its success no doubt inspired the launch of other quantitative hedge funds that have tried and failed to replicate Medallion’s returns. (Renaissance Technologies declined to comment for this article.)
According to The Man Who Solved the Market, Medallion’s strategy involves holding thousands of short-term positions, both long and short, at any given time. The fund makes high-frequency trades, but has also held positions for up to one or two weeks, per Zuckerman’s description. Robert Mercer, the former co–chief executive of Renaissance Technologies, allegedly told a friend that Medallion was right 50.75 percent of the time when it came to its millions of trades — adding that “you can make billions that way.”
In simple terms, the Medallion fund reportedly makes money in much the same way that a casino does. The house doesn’t always win — but enough small wins over time can add up to large profits.
“In Medallion’s situation they’re probably not taking larger bets — they’re taking small bets that are all about the same in terms of profitability,” explains Campbell Harvey, a finance professor at Duke University’s Fuqua School of Business. Harvey, one of the professors thanked in the acknowledgements of Cornell’s paper, explains that being right just over half the time could theoretically result in “a lot of money.”
He adds, “If you’re doing potentially hundreds of thousands or millions of trades, even a small amount of profitability per trade turns out to be a big amount.”
By Cornell’s estimation, Medallion’s apparent trading skill would have turned a $100 investment in the fund at the start of 1988 into $398,723,873 by the end of 2018. “In 31 years, Medallion would have turned a $100 investment into a $400 million fortune,” Cornell writes.
By comparison, $100 invested in the stock market at the beginning of 1988 — using the Center for Research in Security Prices’ value-weighted index — would have grown to $1,910 over the same time period, with dividends reinvested. Even if an investor had the ability to perfectly predict stock market returns on a monthly basis — and had invested in Treasury bills during times of stock market underperformance — Cornell asserts that the investor’s $100 starting investment would have grown to only $331,288 over that time frame.
He implies that human knowledge and action could not generate this kind of performance. Dismissing fraud as the explanation, we are left with the possibility that the source of the performance is, to use the term of the 16th-century Catholic theologians, “not truly human.”
But non-human investing simply is not possible within our current ideology. Non-human investing and, more generally, non-human knowledge require an alternate worldview.
That worldview does exist. It springs from what DeepMind CEO Demis Hassabis calls a “new wave of AI [artificial intelligence] systems.”
He writes: “Traditional AI programs operate according to hard-coded rules [created by humans], which restrict them to working within the confines of what is already known. But a new wave of AI systems, inspired by neuroscience, are capable of learning on their own from first principles.”
These systems do not use human data or human expertise in any fashion — and, precisely because of that, they are not encumbered by those constraints. They are “able to create knowledge itself.”
Other knowledge-based verticals (e.g., medicine) have used these new systems (e.g., deep learning and deep reinforcement learning) to achieve superhuman results — results that exceed those that can be achieved with human knowledge and expertise.
Even as many investors might acknowledge that these models could achieve such superhuman results within narrowly defined applications (such as playing board games and reading CAT scans), they dismiss the possibility that the same systems could be used successfully to build self-learning investment processes by claiming that markets are simply too noisy and too unpredictable.
Not only are these claims unsupported by any empirical evidence — they are grounded in a distinctly human-centric view of markets. After all, “noise” and “predictability” are human constructs.
This argument against an alternative worldview reminds me of Stephen Jay Gould’s remark that “we modern scholars often treat our professions as fortresses and our spokespeople as archers on the parapets, searching the landscape for any incursion from an alien field.”
Yet we live in a time when the defense of our beliefs threatens our future — client dissatisfaction with the current iteration of active management is so great that PwC estimates that “by 2025, at least 20% of fund managers will be acquired or eliminated.”
De qualquer forma deixa que chame a atenção que no meio de retornos desgraçados que este ano naturalmente tem havido no meio da indústria dos hedge-funds aparecem sempre alguns poucos resultados de abrir a boca de espanto.
Vem isto a propósito dos retornos obtidos pelo fundo Universa Investments, um fundo do tipo "tail hedge fund" com o principal objectivo de segurar a carteira em ambientes tipo crash, gerida pelo gestor profissional Mark Spitznagel (mais um matemático graduado em New York, que agora parecem estar na moda!) a obter o valor excepcional de 4144% à data de 7 de abril, incluindo 3612% só no mês de março, um mês extraordinário para quem shortou em cheio nessa altura!
Antes de terminar achei muito curiosa, através dum excerto dum artigo existente no site da Attic Capital, a forma como o Fundo Medallion funciona. Achei espantoso, cada vez que leio novas notícias deste fundo fico de boca aberta, diz-se que a programação das ordens de compra e venda implicam mais de 1 milhão de linhas de relação programática baseadas em complexos algoritmos de matemática e estatística quântica de Inteligêngia Artificial que eliminam toda a emoção humana, que possa eventualmente estar envolvida, e a longo prazo trata-se provavelmente dum veículo de investimento que não tem comparação com algo semelhante que possa existir no mercado:
Inside Renaissance Technologies Medallion Fund
by invest
With a net worth of over $20 billion Jim Simons is known as one of the smartest investors to ever take on wall street. There’s more to Simons beyond his net worth and astounding hedge fund track record. Jim Simons may be one of the most interesting people in business you will find. He’s a mathematician who realized pattern recognition can be applied to trading in financial markets. A recently released book titled, ‘The Man Who Solved the Market: How Jim Simons Launched the Quant Revolution’ by author Gregory Zuckerman details how Simons transformed markets over the last 30-plus years.
RenTec
Renaissance Technologies, also known as ‘RenTec’, is a hedge fund based in New York founded by Jim Simons in 1982. By 1978 Simons grew tired of running the Department of Mathematics at Stony Brook University and began trading in the markets. He set up a hedge fund called Monemetrics in a Long Island strip mall and four years later changed the name of the firm to Renaissance Technologies.
The firm is known for it’s incredible returns over the last three decades and the mystery of how it’s obtained such outstanding performance. How do you get returns unlike any other hedge fund in existence? By doing things completely different than your competition. RenTec hires mathematicians, statisticians, physicists and signal processing experts without financial backgrounds. The company is run by scientists for quantitative research where Wall Street experience is frowned on. RenTec employees are believed to comment that the heard-like mentality among business school graduates is to blame for poor returns in the investment world.
The secrecy surrounding RenTec requires it’s employees to sign lengthy non-disclosure agreements and non-compete documents. Their ability to attract the top scientific minds in the world is a known fact. Mathematician Isadore Singer commented on RenTec’s headquarters as the best physics and mathematics department in the world. Not only does the firm attract brilliant minds but their turnover rate is nearly zero.
The Fund
The Medallion Fund is widely known as one of the best performing funds of all time. It trounces Warren Buffet over many decades. Peter Lynch, Ray Dalio and George Soros all trail in average annual returns. The funds worst year return was a 21 percent gain from 2001 to 2013. Extraordinary results over that long of a time period. Especially considering there was a global financial crisis in the middle of that 12 year period. Not only that but the fund has averaged a 71% return before fees from 1994 to 2014.
You might be asking yourself how to gain access to returns like this. The fact is that you can’t. The Medallion Fund has been closed to new investors since 1993 and is only available to employees of the firm.
The fund is referred to as a printing press. Many on Wall Street have difficulty explaining how the enormous returns are accomplished. Studies show that Medallion’s returns are uncorrelated with many major financial market indexes. Momentum, volatility, small caps, distressed debt, none of these categories indicate a relation to the funds returns.
Another factor to the success of the fund is the inability for any other fund to duplicate it’s strategies. Often times when a hedge fund or hedge fund manager develops a successful track record the methods are quickly traced by competing firms and eventually kills the strategy. This has not happened yet and is a testament to the secrecy of the firms employees inside Renaissance Technologies Medallion Fund.
Deixo abaixo o espectacular gráfico da evolução da estonteante Equity Line do Medallion Fund no período de 1984 a 2003, algo que só pensamos que existe na teoria mas que aqui era bem real!
Finalmente recolhi numa fonte do youtube numa reportagem sobre o Jim Simons, vale o que vale, um pequeno resumo dos 4 pontos chave que estão por trás da estratégia do Medallion Fund:
1) Análise Técnica estatisticamente válida e significativa;
2) Reversão para a média;
3) Utilização exaustiva de bancos de dados;
4) Trading de curto prazo.
BN
Sim, é verdade que o Medallion Fund foi fechado à subscrição do público ao fim de 11 anos em 1993, na altura a gerir perto de 10 biliões USD. Os retornos eram (e continuam a ser!) excepcionais e os pedidos de adesão eram muito elevados, provavelmente foi o motivo para ser fechado aos clientes externos mas outra razão pode estar relacionada com o facto de que é cada vez mais difícil garantir retornos de excelência à medida que o valor dos investimentos a gerir sobem para patamares muito mais elevados. Basta pensar: porque motivo o fundo principal de 200 biliões USD da Berkshire Hathaway do Warren Buffett nunca mais conseguiu bater o mercado? Porque o fundo com aquela dimensão é quase o próprio mercado!
De qualquer maneira os fees no Medallion para os amigos e colaboradores são de 5% "front management fee" do total investido na data de subscrição e de 44% "performance return fee" sobre os lucros obtidos no resgate, cerca de metade fica na casa gestora e quem resgata com o resto fica milionário!
Por curiosidade existe uma entrevista muito interessante no MIT com o Jim Simons, com duração de mais de uma hora, que pode ser vista aqui abaixo para quem tenha tempo livre disponível, de certeza que vai dar por bem empregue o tempo gasto:
https://www.youtube.com/watch?v=srbQzrtfEvY
Ao pesquisar sobre a performance actual dos fundos da Ren Tech do Jim Simons neste ano de 2020 encontrei este curioso artigo desta semana no Institutional Investor:
The Famed Medallion Fund
Is Crushing It. Other
RenTech Funds, Not So Much.
The disparity — a performance difference of 17 to 19 percentage points — is “really surprising,” according to Renaissance Technologies skeptic Bradford Cornell, a professor emeritus at the University of California Los Angeles.
Amy Whyte
April 21, 2020
Not even a crisis can bring down Renaissance Technologies’ market-leading Medallion fund. However, the same cannot be said for the quantitative firm’s other hedge funds — the ones meant largely for outside investors.
According to the Wall Street Journal, the famous — and famously secretive — Medallion fund was up a stunning 24 percent year-to-date through April 14. This included a 9.9 percent gain in the month of March, when stock prices plunged.
Yet Renaissance Technologies’ other funds have not fared so well in 2020. Even after bouncing back in the first half of April, the three funds — Renaissance Institutional Equities Fund, Renaissance Institutional Diversified Alpha, and Renaissance Institutional Diversified Global Equities Funds — were each down between 7 and 9 percent for the year through April 17.
The disparity — a performance difference of some 17 to 19 percentage points — is “really surprising,” according to Bradford Cornell, a professor emeritus at the University of California Los Angeles.
The professor’s surprise is, perhaps, not surprising at all.
Cornell previously analyzed the performance of the Renaissance Technologies funds in a brief paper entitled “Medallion Fund: The Ultimate Counterexample?” In the paper, he wrote that the performance of the Medallion fund — which between 1988 and 2018 delivered a gross annualized return of 66 percent — was “extraordinary.”
Cornell writes in the paper’s abstract: “The performance of Renaissance Technologies’ Medallion fund provides the ultimate counterexample to the hypothesis of market efficiency. To date, there is no adequate rational market explanation for this performance.”
Turning to the returns of the REIF and RIDA funds, however, Cornell found that they were “relatively mundane and in no way comparable to Medallion.”
Wall Street Journal reporter Gregory Zuckerman, the author of The Man Who Solved the Market, says the large gap in returns can be explained by the differences in strategies among the four funds.
“Discrepancies between Medallion's returns and those of other Renaissance funds catch the eye, but I've never seen evidence something nefarious is going on — neither has the SEC, which spent two years in their offices after the Madoff scandal,” he said in an emailed statement. “Medallion has short-term holding periods, RIEF and the other funds search for longer-term aberrations and own smaller stocks and other investments — things Medallion shies away from.”
According to fund documents, the Medallion fund employs a short-term, quantitative trading strategy across multiple asset classes. These include global equities, futures, commodities, and currencies, according to a person familiar with the fund. It also tends to have high turnover and significant leverage.
The Renaissance Institutional Equities Fund, by comparison, only trades in equities, and holds stocks for long periods of time, according to fund’s registration document.
As Institutional Investor previously reported, the REIF fund was created to generate gross annual returns of 400 to 600 basis points — or 4 to 6 percentage points — above the S&P 500 over rolling three- to five-year periods.
The fund was up 8 percent for the month as of April 17. Year-to-date, it was down 7 percent — roughly 400 basis points better than the S&P 500, which was down by over 11 percent for the year through April 17.
The Renaissance Institutional Diversified Alpha fund, meanwhile, trades equities, derivatives, and various instruments in the global futures and forwards markets, according to fund documents. Like REIF, the RIDA fund holds significant individual positions, usually for long periods of time.
It was up 1.55 percent for the month through April 17, for a year-to-date loss of 9 percent.
The Renaissance Institutional Diversified Global Equities Funds, which trades equities and derivatives, was also in the red as of April 17, despite a 3 percent gain for the month thus far. According to fund documents, the RIDGE fund seeks to be market neutral by maintaining low levels of beta, or exposure to the broader market.
But as the firm admitted in the fund registration document, “the beta models in recent volatile markets have not performed as expected.”
Overall, the returns are in stark contrast to that of the Medallion fund — a fund that has been closed to external capital since 1993.
“If they have such secret sauce, how could the public funds be down so much?” Professor Cornell said in an email. “I wish I had an answer.”
Explosively, professor Bradford Cornell’s recent brief analysis of the performance of Renaissance Technologies’ Medallion fund states that “to date, there is no adequate rational market explanation for this performance.”
I find this conclusion interesting — not because of its contribution to our understanding of the Medallion fund’s performance (such efforts have devolved into a parlor game), but because it accidentally reveals an implicit worldview that, upon reflection, is a root cause of the dismal record of active management.
This belief rests on the conceit that investing is essentially a human activity — and therefore human reason must be able to explain the results of such activity.
There is a continuum of expressions of this worldview, ranging from the dogmatic assertion that active investing must be based on either cash flow analysis or short-term price movement to authoritarian claims that there can be no investment activity without human talent.
Through dictatorial educational programs and a homogeneous workforce, we have reached a point where this perspective is unquestionably accepted. This acceptance allows us to pretend that everything worth knowing is already known and in use, making any alternative unimaginable.
In investing, we, too, have our own “scripture” that grounds our perspective, and, though not revelatory, it has been unshaken for the past 60 years.
For example, since Michael Jensen’s 1968 landmark study of the performance of 115 mutual fund managers, we have held the view that active management cannot consistently provide returns better than a relevant passive proxy.
Yet the “extraordinary” and “unprecedented” performance of the Medallion fund presents us with a phenomenon that contradicts this fundamental belief. “In forty plus years of reading hundreds of papers on investment anomalies, including some that benefited from data snooping and ex-post selection bias, I have never seen any performance approaching that reported by Medallion,” professor Cornell concluded in his paper.
Ultimately, Cornell wrote that he could come up with no “convincing” explanation for the Medallion fund’s outsize returns, noting that even if Medallion was simply better at trading than any other fund, “the returns are so large, it stretches that explanation to the limit.”
Cornell’s incredulity is not an unusual reaction to the Medallion fund, which has long baffled industry insiders and observers. The secret to Renaissance Technologies’ performance has been debated in news articles and academic circles and on online message boards — and its success no doubt inspired the launch of other quantitative hedge funds that have tried and failed to replicate Medallion’s returns. (Renaissance Technologies declined to comment for this article.)
According to The Man Who Solved the Market, Medallion’s strategy involves holding thousands of short-term positions, both long and short, at any given time. The fund makes high-frequency trades, but has also held positions for up to one or two weeks, per Zuckerman’s description. Robert Mercer, the former co–chief executive of Renaissance Technologies, allegedly told a friend that Medallion was right 50.75 percent of the time when it came to its millions of trades — adding that “you can make billions that way.”
In simple terms, the Medallion fund reportedly makes money in much the same way that a casino does. The house doesn’t always win — but enough small wins over time can add up to large profits.
“In Medallion’s situation they’re probably not taking larger bets — they’re taking small bets that are all about the same in terms of profitability,” explains Campbell Harvey, a finance professor at Duke University’s Fuqua School of Business. Harvey, one of the professors thanked in the acknowledgements of Cornell’s paper, explains that being right just over half the time could theoretically result in “a lot of money.”
He adds, “If you’re doing potentially hundreds of thousands or millions of trades, even a small amount of profitability per trade turns out to be a big amount.”
By Cornell’s estimation, Medallion’s apparent trading skill would have turned a $100 investment in the fund at the start of 1988 into $398,723,873 by the end of 2018. “In 31 years, Medallion would have turned a $100 investment into a $400 million fortune,” Cornell writes.
By comparison, $100 invested in the stock market at the beginning of 1988 — using the Center for Research in Security Prices’ value-weighted index — would have grown to $1,910 over the same time period, with dividends reinvested. Even if an investor had the ability to perfectly predict stock market returns on a monthly basis — and had invested in Treasury bills during times of stock market underperformance — Cornell asserts that the investor’s $100 starting investment would have grown to only $331,288 over that time frame.
He implies that human knowledge and action could not generate this kind of performance. Dismissing fraud as the explanation, we are left with the possibility that the source of the performance is, to use the term of the 16th-century Catholic theologians, “not truly human.”
But non-human investing simply is not possible within our current ideology. Non-human investing and, more generally, non-human knowledge require an alternate worldview.
That worldview does exist. It springs from what DeepMind CEO Demis Hassabis calls a “new wave of AI [artificial intelligence] systems.”
He writes: “Traditional AI programs operate according to hard-coded rules [created by humans], which restrict them to working within the confines of what is already known. But a new wave of AI systems, inspired by neuroscience, are capable of learning on their own from first principles.”
These systems do not use human data or human expertise in any fashion — and, precisely because of that, they are not encumbered by those constraints. They are “able to create knowledge itself.”
Other knowledge-based verticals (e.g., medicine) have used these new systems (e.g., deep learning and deep reinforcement learning) to achieve superhuman results — results that exceed those that can be achieved with human knowledge and expertise.
Even as many investors might acknowledge that these models could achieve such superhuman results within narrowly defined applications (such as playing board games and reading CAT scans), they dismiss the possibility that the same systems could be used successfully to build self-learning investment processes by claiming that markets are simply too noisy and too unpredictable.
Not only are these claims unsupported by any empirical evidence — they are grounded in a distinctly human-centric view of markets. After all, “noise” and “predictability” are human constructs.
This argument against an alternative worldview reminds me of Stephen Jay Gould’s remark that “we modern scholars often treat our professions as fortresses and our spokespeople as archers on the parapets, searching the landscape for any incursion from an alien field.”
Yet we live in a time when the defense of our beliefs threatens our future — client dissatisfaction with the current iteration of active management is so great that PwC estimates that “by 2025, at least 20% of fund managers will be acquired or eliminated.”
De qualquer forma deixa que chame a atenção que no meio de retornos desgraçados que este ano naturalmente tem havido no meio da indústria dos hedge-funds aparecem sempre alguns poucos resultados de abrir a boca de espanto.
Vem isto a propósito dos retornos obtidos pelo fundo Universa Investments, um fundo do tipo "tail hedge fund" com o principal objectivo de segurar a carteira em ambientes tipo crash, gerida pelo gestor profissional Mark Spitznagel (mais um matemático graduado em New York, que agora parecem estar na moda!) a obter o valor excepcional de 4144% à data de 7 de abril, incluindo 3612% só no mês de março, um mês extraordinário para quem shortou em cheio nessa altura!
Antes de terminar achei muito curiosa, através dum excerto dum artigo existente no site da Attic Capital, a forma como o Fundo Medallion funciona. Achei espantoso, cada vez que leio novas notícias deste fundo fico de boca aberta, diz-se que a programação das ordens de compra e venda implicam mais de 1 milhão de linhas de relação programática baseadas em complexos algoritmos de matemática e estatística quântica de Inteligêngia Artificial que eliminam toda a emoção humana, que possa eventualmente estar envolvida, e a longo prazo trata-se provavelmente dum veículo de investimento que não tem comparação com algo semelhante que possa existir no mercado:
Inside Renaissance Technologies Medallion Fund
by invest
With a net worth of over $20 billion Jim Simons is known as one of the smartest investors to ever take on wall street. There’s more to Simons beyond his net worth and astounding hedge fund track record. Jim Simons may be one of the most interesting people in business you will find. He’s a mathematician who realized pattern recognition can be applied to trading in financial markets. A recently released book titled, ‘The Man Who Solved the Market: How Jim Simons Launched the Quant Revolution’ by author Gregory Zuckerman details how Simons transformed markets over the last 30-plus years.
RenTec
Renaissance Technologies, also known as ‘RenTec’, is a hedge fund based in New York founded by Jim Simons in 1982. By 1978 Simons grew tired of running the Department of Mathematics at Stony Brook University and began trading in the markets. He set up a hedge fund called Monemetrics in a Long Island strip mall and four years later changed the name of the firm to Renaissance Technologies.
The firm is known for it’s incredible returns over the last three decades and the mystery of how it’s obtained such outstanding performance. How do you get returns unlike any other hedge fund in existence? By doing things completely different than your competition. RenTec hires mathematicians, statisticians, physicists and signal processing experts without financial backgrounds. The company is run by scientists for quantitative research where Wall Street experience is frowned on. RenTec employees are believed to comment that the heard-like mentality among business school graduates is to blame for poor returns in the investment world.
The secrecy surrounding RenTec requires it’s employees to sign lengthy non-disclosure agreements and non-compete documents. Their ability to attract the top scientific minds in the world is a known fact. Mathematician Isadore Singer commented on RenTec’s headquarters as the best physics and mathematics department in the world. Not only does the firm attract brilliant minds but their turnover rate is nearly zero.
The Fund
The Medallion Fund is widely known as one of the best performing funds of all time. It trounces Warren Buffet over many decades. Peter Lynch, Ray Dalio and George Soros all trail in average annual returns. The funds worst year return was a 21 percent gain from 2001 to 2013. Extraordinary results over that long of a time period. Especially considering there was a global financial crisis in the middle of that 12 year period. Not only that but the fund has averaged a 71% return before fees from 1994 to 2014.
You might be asking yourself how to gain access to returns like this. The fact is that you can’t. The Medallion Fund has been closed to new investors since 1993 and is only available to employees of the firm.
The fund is referred to as a printing press. Many on Wall Street have difficulty explaining how the enormous returns are accomplished. Studies show that Medallion’s returns are uncorrelated with many major financial market indexes. Momentum, volatility, small caps, distressed debt, none of these categories indicate a relation to the funds returns.
Another factor to the success of the fund is the inability for any other fund to duplicate it’s strategies. Often times when a hedge fund or hedge fund manager develops a successful track record the methods are quickly traced by competing firms and eventually kills the strategy. This has not happened yet and is a testament to the secrecy of the firms employees inside Renaissance Technologies Medallion Fund.
Deixo abaixo o espectacular gráfico da evolução da estonteante Equity Line do Medallion Fund no período de 1984 a 2003, algo que só pensamos que existe na teoria mas que aqui era bem real!
Finalmente recolhi numa fonte do youtube numa reportagem sobre o Jim Simons, vale o que vale, um pequeno resumo dos 4 pontos chave que estão por trás da estratégia do Medallion Fund:
1) Análise Técnica estatisticamente válida e significativa;
2) Reversão para a média;
3) Utilização exaustiva de bancos de dados;
4) Trading de curto prazo.
BN