Broken Markets: How High Frequency Trading and Predatory Practices on Wall Street are Destroying Investor Confidence and Your Portfolio (Chris Mathews' Library) by Sal Arnuk & Joseph Saluzzi

Author:Sal Arnuk & Joseph Saluzzi
Language: eng
Format: epub
Publisher: FT Press
Published: 2012-06-27T16:00:00+00:00

How Algorithms Work: Making the Sausage

A portfolio manager (PM) generates the order, which then shows up on his trader’s order management system (OMS). The PM explains his game plan to his trader. He can tell the trader he is passive and just quietly adding to a position, or that he is aggressive and wants to buy his stock quickly. The trader decides whether to allow the order to be seen by electronically connected “scraping” dark pools, such as Liquidnet. If he does allow such a dark pool to see his order, his order may trade with a large order on the opposite side of his. Having “blocked” a portion of his order, or not, he can next choose among several options. He can send the order to a broker whose capabilities he trusts, such as Themis Trading, after explaining his game plan. He also can choose to work an order manually, feeding portions to exchanges and dark pools of his choosing. Or he can send an order into any number of the broker-sponsored algorithmic suites at his disposal.

Typically, within each broker’s suite are several variations of the following types of algos:

• Volume weighted average price (VWAP): Breaks the parent order into smaller child orders that are sent into the markets throughout the day at a rate based on the stock’s historical intraday volume dispersion.

• Time weighted average price (TWAP): Breaks the parent order into smaller child orders that are fed into the market at equal time increments throughout the day.

• Percentage of volume (POV): Delivers child orders into the market at trader-defined volume participation rates, such as a “10% of the trading volume order.”

• Close-targeting-algos: Seeks to “beat” the day’s closing price.

• Arrival-targeting-algos: Seeks to beat the price of the stock when the algo first starts working.

• Dark liquidity seeking algos: Looks for liquidity away from the public stock exchanges.

• “Next Generation” algos: Typically a variation of the above algos, using what’s billed as “anti-HFT methodology.”

If the trader chooses to send the order to an algorithm himself, the algo selected depends on the PM’s goal for that specific order. For example, if there is a conference at which a company’s management will be speaking midday, the PM might believe that management will make positive comments on business prospects. The trader might want to buy all the stock quickly, so he might use an algo focused on arrival price or one that targets a high percentage of volume. Conversely, if the PM expects no news in the short term and just wants to add quietly to his position, his trader might use a less aggressive option, such as a low POV algo or a constrained VWAP algo with price limits.

After the trader chooses an algorithm, he loses control over the process. Every broker’s algorithms interact differently with the more than 50 exchanges and dark pools. Algos vary widely in where they route, in what order, and with what specific constraints. What happens to the institutional trader’s order can be visually described in the diagram in Figure 9.

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.