why does it matter to anyone if a human or a computer is on the other side of a trade? if the market is mispriced someone will correct. and the guy clicking a mouse at home benefits many people: his broker, he makes/takes trades. the programs that trade just make the markets more efficient and as a programmer I always like to say "I don't make people's jobs easier I make them nonexistent"
Really? So an algo can adjust on the fly? Not being facetious, just wanting to learn what I'm up against is all. The biggest advantage of us is being able to read and react, or if you will, the artistic side of trading. Now if you're telling me that algos can take human emotion into account and change their "rithm" in mid-step, then I would imagine we are in trouble.
as an example, a MM algo could monitor N period volatility and if it gets too high it could widen out its market, or if it gets too long/short it could adjust parameters. or if large volume comes in it could adjust. basically what a human does but on a consistent basis
Algorithms can be as flexible as their programmers want them to be. There's nothing out there that can't be converted to an algorithm. Consider time-series models like ARMA-GARCH. Computers don't have to be so dumb as to have strategies that are invariant to market volatility. By introducing a model that revolves around heteroscedascity, you've got something market enough to adapt to shifting volatilities and can function in a wide number of cases.
OK, I understand. But it's still rule based isn't it? If y does this, then x does this to produce (results)... sorry I'm not versed in math nomenclature. I'm sure with a powerful enough program you could formulate everything (emotion?) into X's and 0's but are we there yet? I don't think so. From what I've read quants have been getting their arse handed to them. I just think as prevalent as these programs are, a good intuitive trader has a chance to turn it as an advantage as long as the liquidity is there. There's room for everybody to fail.
Speaking of which, here is an interesting article on Zero Hedge. http://zerohedge.blogspot.com/2009/05/observations-on-nyse-program-trading.html Now I'm all for capitalism in it's purest form, whatever that means, but it seems that GS's dick is getting just way to big for it's body. We're talking a financial universe worth trillions and seems that Goldies finger is in every conceivable hole. Now cornering the market of SLP's? That means either no one wants a piece of that pie, or Goldie wants the whole pie for itself.
When 1-2 or even a handful of big operators like GS are allowed to control such high volumes, the market becomes manipulated. A lot of sector moves lately are totally fake BS and this kind of thing partly explains why! It seems the NYSE, in order to fill their own pockets, encourages/pays these crooks to play games with prices so that others are sucked into the game to "add liquidity". I'm trading other markets outside US more than ever. All I see is manipulated crap.
when has this game ever been anything but 10% picking off the other 90%? So the technology and players change, and there will always be fresh meat as long as markets exist. The high fequency people will just start picking each other off sooner or later.
Few interesting books coming out in this area: Chasing the Same Signals: How Black-Box Trading Influences Stock Markets from Wall Street to Shanghai --------------------------------------------- "The worst stock market crash since Black Monday during October of 1987 occurred during the first week of August of 2007. But nobody noticed. On the morning of August 6th 2007, investment professionals were baffled with unprecedented stock patterns. Mining sector stocks were up +18% but manufacturing stocks were down -14%. It was an extreme sector skew yet the S&P index was unchanged at +0.5% on the day. The next few days would continue with excessive volatility. MBI Insurance, a stock that had rarely attracted speculation would finish up +15% on Aug 6th, followed by another +7% on Aug 7th, and then finish down -22% over the subsequent two days. The brief rally in MBI was short lived. Only weeks later would investors begin to have insights on the dispersion patterns. Prominent hedge funds that had never had a negative annual performance began disclosing excessive trading loses with many notable firms reporting several hundred millions were lost - in a single day. Hedge funds were hemorrhaging in excess of 30% of their assets when the S&P index was unchanged. The market dispersion was the side effects of the synchronous unwind ignited by the hordes of 'computerized' strategies that were caught off guard when history didn't repeat. It was the industry's first world wide panic - by machines. Over the past decade, computerized (or black-box) trading has had a coming of age. Black-box firms use mathematical formulas to buy and sell stocks. The industry attracts the likes of mathematicians, astrophysics and robot scientists. They describe their investment strategy as a marriage of economics and science. Their proliferation has been on the back of success, black-box firms have been among the best performing funds over the past decade, the marquee firms have generated double-digit performance with few if any months of negative returns. Through their coming of age, these obscure mathematicians have joined the ranks of traditional buy-n-hold investors in their influence of market valuations. A rally into the market close is just as likely the byproduct of a technical signal as an earnings revision. They are speculated to represent a one third of all market volume albeit their influence to the day-to-day gyrations goes largely unnoticed. CNBC rarely comments on the sentiments of computerized investors. Conventional wisdom suggests that markets are efficient, random walks and that stock prices rise and fall with the fundamentals of the company. How then have black-box traders prospered and how do they exploit market inefficiencies? Are their strategies on their last legs or will they adapt to the new landscape amidst the global financial crisis? Chasing the Same Signals is a unique chronicle of the black-box industry's rise to prominence and their influence on the market place. This is not a story about what signals they chase, but rather a story on how they chase and compete for the same signals." --------------------------------------------- Chapter 1 Introduction. DoCoMo Man. Rules Based Trading. Our Economic Barometer. Chapter 2 The Black Box Investment Philosophy. The Black Box Community. The Thought Process. The Investment Objective. Chapter 3 An Adaptive Industry. Period of Soul Searching. Behavioral Economics. Natural Selection. Chapter 4 Finding the Footprint. Momentum Trading. Demographics of Liquidity. The Shortfall Movement. Chapter 5 Disciples of Risk Factors. Market Neutral Strategies. Quantitative Stock Selection. A Leveraged Life. Chapter 6 The Game of High Frequency. Pining the Book. Liquidity Providers. Automated Specialists. Chapter 7 Era of Execution Strategies. The Russell Rebalance. Sourcing Liquidity. Risk Consolidation. Chapter 8 Globalization of Equity Markets. The Microstructure Layer. Dispersion Opportunities. Market Structure. Chapter 9 A Future of Adaptation. The Man Machine Interface. Adaptive Market Place. Outlook for the Species. -------------------------------------------- Narang, Rishi K Inside the Black Box A straightforward look at quantitative trading Investors, from high-net-worth individuals to pension funds, have never been more interested in quantitative trading-mainly due to the impressive returns they usually generate. And yet, few actually understand what goes on inside these black box trading strategies. That's why expert fund manager Rishi Narang has created Inside the Black Box. In non-mathematical terms-and supplemented by anecdotes and real-world stories-this guide explains how quantitative trading strategies actually work. Written in a straightforward and accessible style, this book also skillfully explains how quant strategies fit into a portfolio, why they are valuable, and how to evaluate a quant manager. Some of the questions covered throughout these pages include: How do quants capture alpha? What is the difference between theory-driven systems vs. data-mining strategies? How do quants model risk? * Demystifies quantitative trading and how it works * Provides key information that investors need to evaluate the best quantitative trades * Explains the essential elements of this discipline without heavy-handed mathematical jargon For anyone looking to gain a better understanding of quantitative and algorithmic trading, Inside the Black Box is a highly recommended read. Foreword. Acknowledgments. Chapter 1: Why Does Quant Trading Matter? The Measurement and Mismeasurement of Risk. Summary. What is a Quant? Summary. Chapter 3: How Do Quants Make Money? Types of Alpha Models: Theory-driven and Data-driven. Data-Driven Alpha Models. Blending Alpha Models. Chapter 4: Risk Models. Limiting the Types of Risk. Chapter 5: Transaction Cost Models. Types of Transaction Cost Models. Chapter 6: Portfolio Construction Models. Portfolio Optimizers. How Quants Choose a Portfolio Construction Model. Chapter 7: Execution. High Frequency Trading: Blurring the Line between Alpha and Execution. Summary. The Importance of Data. Sources of Data.. Storing Data. Chapter 9: Research. Idea Generation. Summary. Chapter 10: Risks Inherent to Quant Strategies. Regime Change Risk. Contagion, or Common Investor, Risk. Summary. Setting the Record Straight. Quants Cause More Market Volatility by Underestimating Risk. Quants are All the Same. Quants are Guilty of Data Mining. Chapter 12: Evaluating Quants and Quant Strategies. Evaluating a Quantitative Trading Strategy. The Edge. How Quants Fit into a Portfolio. Chapter 13: Looking to the Future of Quant Trading. About the Author.
A computer prodigy recently approached me and said he had a new computer architecture that was thousands of times faster than existing CPUs and hundreds of times faster than existing GPUs. The guy's legit. Leading researcher in artificial intelligence. Also a Phd professor at a fancy university. I don't know the first thing about the guts of a computer, but love high frequency trading - where algos fight each other for my order in microsecond time frames (no longer milliseconds per the original post). Only if you're serious, and have serious connections in the finance or hardware industry, PM me with how you think this guy should proceed.