I believe this is the leading algo platform in the industry so this would be a start. Knowing the technology would be helpful. http://www.progress.com/realtime/algorithmic_trading/index.ssp Algorithmic trading is a technique used by institutional investors to break up large trades into smaller blocks to avoid detection by other traders, floor specialists and market makers. It is also the latest and greatest form of computerized financial trading, pioneered by hedge funds. The Progress Real Time Division (PRTD) provides modeling tools, including Progress® Apama® ATP (Algorithmic Trading Platform), that greatly shorten the time to create and deploy new trading strategies so traderâs can constantly change and evolve their business requirements. In terms of flexibility and ease of strategy development and modification, the Apama ATP (Algorithmic Trading Platform) environment is "white box" rather than "black box". Traders can modify strategies at any time, thus leveraging their own scenarios and intellectual property to create ever-evolving financial trading applications that outperform commodity trading "black boxes." And with its robust event stream processing platform, Apama's ATP-based strategies can monitor unprecedented levels of market activity and respond instantly-giving traders the crucial first-mover advantage that opportunistically capitalizes on new or changing conditions with sophisticated trading actions that are automatically spawned from the algorithmic trading platform. "Hedge Funds Increasing competition between Hedge Funds is causing tremendous interest in algorithmic trading. Apama enables users to customize strategies to incorporate the Hedge Fund's secret ingredients, with the strategies deployed into a live market and traded instantly. For funds with an interest in "high frequency trading" and algorithmic and quantitative trading, Apama enables models to be built rapidly and evolved easily, so that new opportunities can be capitalized upon effectively. And the out-of-the-box Application Modules for popular strategies such as VWAP, pairs trading and index arbitrage provide a great starting point. Hedge Funds also want to benefit from Direct Market Access (DMA): the ability to use a broker for market access but manage your own execution, thus radically reducing cost. Apama can be used as a platform for DMA-based algorithmic trading. " "Buy-Side Advantage Increasing competition between trading institutions is generating considerable interest in algorithmic trading because of the opportunities for competitive advantage it offers. Rather than deploy broker-provided "black box" strategies, Apama's integrated environment and robust tools can be installed directly into the buy-side firm's environment, enabling those institutions to develop unique, differentiated strategies. With Apama, strategies can be developed and evolved quickly, deployed in a live market and traded instantly with any number of "strategy instances" run concurrently. Apama's advanced development and execution environment enables proprietary traders to create and deploy algorithmic trading strategies. Traders can use Apama's graphical tools to rapidly build reusable analytics which are incorporated into multiple different trading strategies. And traders who wish to program directly can opt to code their trading strategies in a concise scripting language or Java. Buy-Side organizations can opt to take Apama in-house or access Apama through a partner, such as a broker."
Look at this 1 year chart for the black box and white box firm. Hmmm, I wonder if such strategies and technology are picking up steam. LOL 7:35am 06/16/05 Progress Software Q2 profit rises 51% (PRGS) By Leslie Wines NEW YORK (MarketWatch) - Progress Software Inc. (PRGS) Thursday reported second-quarter net income of $12.2 million, or 30 cents a share, up 51% from $8 million, or 21 cents a share, a year earlier, citing higher sales and software license profits. On an adjusted basis, earnings were 35 cents a share, up from 24 cents a share, a year earlier. The Thomson First Call analysts average estimate was for earnings of 31 cents a share. Revenue rose 10% to $100.2 million. The Bedford, Mass maker of application infrastructure also forecast third-quarter earnings of 24 cents to 26 cents a share, and of 30 cents to 32 cents a share, on an adjusted basis; the First Call estimate is for quarterly earnings of 31 cents a share. For 2005 Progress Software predicted adjusted earnings of $1.26 to $1.30 a share, with revenue of $397 million to $403 million. The First Call estimate is for earnings of $1.24 a share. On Wednesday the company's stock fell 25 cents to $28.20.
"The market is being put to the test continually by (...) the floor traders. These shrewd fellows are always on the alert to ferret out a weak spot in the market..." (Richard Wyckoff, Studies in Tape Reading, 1910) Big names like Marty Schwartz and Victor Sperandeo openly worried about the rise of program trading, so there is no shame in worrying as well. What did they do? They adapted their methodologies where they saw fit and moved on++
http://www.nyse.com/Frameset.html?displayPage=/press/1120128532486.html http://www.nyse.com/pdfs/pt062205.pdf Program Trading Averaged 76.3 Percent of NYSE Volume during June 20-24 View the Top 15 Most Active (pdf format) NEW YORK, June 30, 2005 -- The New York Stock Exchange today released its weekly program-trading data submitted by its member firms. The report includes trading in all markets as reported to the NYSE for June 20-24. The data indicated that during June 20-24, program trading amounted to 76.3 percent of NYSE average daily volume of 1,751.0 million shares, or 1,335.5 million shares a day. Program trading encompasses a wide range of portfolio-trading strategies involving the purchase or sale of a basket of at least 15 stocks with a total value of $1 million or more. Program trading is calculated as the sum of the shares bought, sold and sold short in program trades. The total of these shares is divided by total reported volume. This is not the only way to measure program trading. Three alternatives for June 20-24 would be to: examine buy programs as a percentage of total purchases (37.8 percent); examine sell programs as a percentage of total sales (38.5 percent); examine program purchases and sales as a percentage of total purchases and sales (38.1 percent); For the second quarter of 2005, the NYSEâs program-trading report includes profiles of program trading whenever the Dow Jones Industrial Average moves 210 points or more in a single direction during any one-hour period. During June 20-24, there were no such periods. In all markets, program trading by member firms averaged 2,859.6 million shares a day during June 20-24. About 46.7 percent of program trading took place on the NYSE, 4.3 percent in non-U.S. markets and 49.0 percent in other domestic markets, including Nasdaq, the American Stock Exchange and regional markets. In aggregate, program volume executed on the NYSE by firms as agent, for non-member customers, amounted to 48.1 percent during June 20-24. Program volume executed as principal, for their own accounts, amounted to 32.7 percent of program volume. Another 19.1 percent was designated as customer facilitation, in which a member firm established or liquidated a principal position to facilitate a program order initiated by a customer. Of the five member firms reporting the most program trading activity on the NYSE, UBS Securities, LLC. executed most of its program trading as principal for its own account. Goldman, Sachs & Co., Citigroup Global Markets, Deutsche Bank Securities and Morgan Stanley & Co. Inc. executed most of their program trading activity for customers, as agent. During June 20-24, 5.5 percent of program volume executed by NYSE member firms related to index arbitrage. Index arbitrage is defined as the purchase or sale of a basket of stocks in conjunction with the sale or purchase of a derivative product such as stock-index futures, to profit from the price difference between the basket and the derivative product. Less than 0.1 percent involved derivative product-related strategies (besides index arbitrage) that are subject to Rule 80A. The rule provides that derivative-related program strategies be executed only in a stabilizing manner after the DJIA moves 210 points or more from the previous dayâs close. In addition to index arbitrage, such strategies include customer facilitations, liquidation of facilitations, index substitutions, liquidation of error accounts, risk modifications, and liquidation of exchange-for-physicals stock positions. All other types of portfolio-trading strategies combined accounted for 94.5percent of member firmsâ program-trading volume during June 20-24.
if you go into the pdf you can see who dominates the market these days. one must remember that it is these fed member banks that got the penny increments in the first place. they are the huge beneficiaries now. i conclude that FX is correct in his thesis. what amazes me now is how hard it is to actually execute a fairly small order these days during a fast market. the exchange and SEC sold us this penny shit under the ruse of helping the little guy....LOL LOL LOL. yeah right. MF'ers
This is a quote from the book "The Money Game" by Adam Smith. It was first published in 1967. This excerpt is from Chapter 12. Don't believe me? Buy the book. "But," Irwin went on, "there are a couple of sophisticated funds that have computers like ours on the air. Then it really gets to be fun. Our computer scans the pattern of the other computer on the air, what its buying and selling programs seem to be. Once we get its pattern, we can have all kinds of fun. We can chase the stock away from it. Or, even better, we can determine where the other computer wants to buy..." Anyway, you get the idea. You've got 40 years of catching up to do if you want to start wrecking the big boys' 'black boxes'. Pointless if you ask me.
How program trading in the NYSE files is defined? Are orders sent through API regarded as program trading? I think orders generated by black boxes account for only a small portion, most are sent by traders assisted with computer programs.
My impression, and I could be mistaken, is that most of this program trading is just big baskets of stocks being bought or sold by computers. The transactions are usually undertaken on behalf of regular client flow business, dispersion books, stat arb types and factor model types. But regardless of who is doing the selling, the algorithims behind the programs are essentially trying to get a higher price for sales and a lower price for buys. I don't know how they determine when is a good time to buy or sell - that might be a good area for research if one wanted to game the algorithms. But generally, it's just a mechanized way of working a client order - dribbling in the order to beat VWAP or some other benchmark. So, it should have a smoothening effect in the intraday I would guess.