Flash Mobs of Trading Robots Coalescing To Rule Markets
An anonymous reader writes "Financial markets experienced a series of computer glitches recently that brought operations to a halt. According to a researcher at the University of Miami, mobs of ultrafast robots, which trade and operate at speeds beyond human capability, may be responsible for these "flash freezes". From the article: '"Even though each trading algorithm/robot is out to gain a profit at the expense of any other, and hence act as a predator, any algorithm which is trading has a market impact and hence can become noticeable to other algorithms," said Neil Johnson, a professor of physics at the College of Arts and Sciences at the University of Miami (UM) and lead author of the new study. "So although they are all predators, some can then become the prey of other algorithms depending on the conditions. Just like animal predators can also fall prey to each other." When there's a normal combination of prey and predators, he says, everything is in balance. But once predators are introduced that are too fast, they create extreme events.'"
This sort of financial activities is complete economic nonsense, as it brings nothing of value to people, companies or other concerns that actually produce something useful to society as a whole. Just reading the /. blurb should be enough to convince anyone that "robot trading" is a parasitic activity that should be taxed to oblivion - by ways of a tax based on the speed of trading for instance - and financial markets forced to become what they're supposed to be: places for investors to invest in real economic activities for the long haul.
"A door is what a dog is perpetually on the wrong side of" - Ogden Nash
Why?
Because we, the people, end up bailing out these irresponsible fuckers who have turned Wallstreet into a casino. The next bailout better be paid in advance by those who caused it, hence a tax on gambling with stocks.
Please, don't disparage the good folks in the probabilistic entertainment area of the hospitality sector by such comparisons.
Casinos may be tacky; and they do suck some gambling addicts dry; but their danger to the larger economy, and to parties who don't choose to deal with them, is quite minimal. Even better, because of their tackiness and the widespread knowledge of how foolish it is to work with them when greater-than-recreational amounts of money are on the line, nobody proposes massive bailouts, or handing social security over to them to manage!
All this news is underlining is that the exchanges are having more of their crumbs stolen by independent parties... if you want reform, start with brining transparent to the stock marker exchanges and their skimming off the top practices. The cost to society is enormous.
No, they don't. They don't have ESP. They can't see what you do on a market before you actually do something on the market.
Actually they do have esp.
They can see what order you have placed before it is fulfilled.
Have you not heard of level 2 market data.
Pay for it and you can see what orders are placed - volume, bid/ask price, trading organisation.
Combine that with super high speed connections to the exchange and you can see transactions at the millisecond and act on them.
Off topic:
If you've ever traded you may have spotted the effects of high frequency when using candlestick charts at the 1 minute timeframe.
You will see long repeated sequences of Bull Candle, Bear Candle of the same size over a short time scale (seconds)
Thats proof of Banks and institutional traders bots / algorithms fighting for price power.
8 Years ago you would be lucky to have spotted a dozen a year.
Nowadays there's one everyday if you know which instrument to look at.
Consider the following quote from the paper
This statement implies that the authors believe a gaussian model "should" apply to the market dynamics. As Benoit Mandelbrot and many others before and after him point out, financial markets never have followed gaussian dynamics and they probably never will. It's especially silly because they go on to analyze the distribution of Ultrafast Extreme Event (UEE) sizes as a power law.
Today's market has both accumulation algorithms now used by mutual funds and other sophisticated "buy and hold" investors, and market-making algorithms used by HF firms, and I fully believe there is some interesting dynamics arising from all that. Whether it is any weirder than the slower, human-derived, dynamics of yesteryear is still in doubt. Humans are so much more complex than any of those algorithms that I suspect if you examined the market behavior in 1980, and sped it up, you would see plots wilder than anything Nanex produces.
The paper is somewhat interesting, but not very convincing.
Being European, I prefer 1789. Off with their heads!