Financial Education

Algorithmic trading systems use electronic trading methods driven by quantitative rules and user-specified benchmarks and constraints. The objective of an algorithmic trading system is to exploit patterns of market trading volume to control execution costs and risks by breaking large trades into smaller, more manageable pieces.

The main types of trading algorithms are logical participation strategies and implementation shortfall strategies.

In a logical participation strategy, one of a number of simple rules is used to determine the proper size of each trade.

• VWAP strategy – uses an estimated VWAP  to break up the order based on a predicted volume profile to match or improve upon the daily VWAP.
• Time weighted VWAP assumes a flat volume profile throughout the day and trades in proportion to time passed. This strategy is useful for illiquid securities.
• Percentage of volume – trades are placed in proportion to the total volume throughout the day until completed.

Implementation shortfall strategies solve for an “optimal” trading strategy, which should minimize the trading costs when measured as implementation shortfall. Such strategies are often front-loaded to take advantage of the higher volume that usually occurs early in the trading day. It can be a useful strategy for portfolio trades or for managing transitions between managers.

The two most common types of orders that can be placed when buying a security are market orders and limit orders.

Market orders are executed promptly at the best available price, which in the case of a buy order is the lowest price a prospective seller is willing to accept. If the buyer wants more shares than the lowest asking party is offering, the remaining shares would be sold to the next-lowest ask price and so on until all shares requested were purchased.

For example, a buyer places a market order for 10,000 shares of stock X. The lowest ask price is offering 5,000 shares at $100 and these are filled. The next lowest ask price is for 3,000 shares at $100.25 and these are filled. The third-lowest ask price is for 5,000 shares at $100.50 and 2,000 of these are filled to complete the order. The effective purchase price is ((5,000 X 100) + (3,000 X 100.25) + (2,000 X 100.50))/10,000 = $100.175

Market orders emphasize prompt execution, and in return result in accepting some uncertainty as to the actual execution price.

Limit orders are instructions to accept only those prices that are better than a designated limit, including a time (end of day, good-til-canceled, etc) at which the order will expire.

For example, the buyer above may specify that the order has a $100.25 limit price. In that case, 5,000 shares would be executed at $100 and 3,000 at $100.25. The others would remain on order until a prospective seller was willing to accept $100.25 or less. The average price at execution would be ((5,000 X 100) + (3,000 X 100.25))/8,000 = $100.09375. If the remaining 2000 shares were executed at $100.25 the effective price would be ((5,000 X 100) + (5,000 X 100.25))/10,000 = 100.125.

Limit orders emphasize price at the expense of uncertainty as to whether the order will be filled in entirety.

Investors typically think of buying stocks that they believe will go up, and selling those (holding no position) in those they believe have poor prospects. Short selling takes this one step further, allowing an investor to have a negative position in a stock they believe will go down in value.

A short sale is accomplished by selling stock that was borrowed from another investor. In exchange for this privilege, the short seller must pay interest on the borrowed stock, as well as any dividends paid by the stock while it is sold short. At an unspecified future point, the short seller buys back the stock and returns it to the original investor.

Short sales are profitable if the stock declines in value while the seller is short, and lose money if the stock increases in value.

This article was originally written by Richard Wilson in his Hedge Fund Blog.

Prime Broker Definition
A large bank or securities firm that provides various administrative, back-office and financing services to hedge funds and other professional investors. Prime brokers can provide a wide variety of services, including trade reconciliation clearing and settlement, custody services, risk management, margin financing, securities lending for the purpose of carrying out short sales, record keeping, and investor reporting. A prime brokerage relationship doesn’t preclude hedge funds from carrying out trades with other brokers, or even employing others as prime brokers. To compete for business, some prime brokers act as incubators for funds, providing office space and services to help new fund managers get off the ground.

Different types of trades require different trading tactics. Some tactics include:

• Liquidity at any cost – need for execution outweighs market impact of trading large blocks
• Need trustworthy agent – large blocks of illiquid securities may require a time/price trade-off and the help of a discreet broker, who of course will require a higher commission
• Costs not important – the need for certain execution justifies paying the full spread, which is at least a competitive price
• Advertise to draw liquidity – for large trades with low information advantage, entering the order between the spread may draw counterparties. Or, it may draw front-runners who drive the price away
• Low cost whatever the liquidity – minimizing commission and trading costs outweighs the risk of failed execution

Traders can be characterized by their motivation to trade.

Information Motivated

Information motivated traders are those who believe they have information whose value declines over time (as it becomes recognized by other market participants.) These types of traders seek speedy execution, and are less sensitive to the price at the time of trade. If successful over time, information traders will want to hide their identity so that potential trading counterparties will not suspect their motive and drive the security price away from them.

Liquidity traders are likely to use market orders, and often rely on market makers to fill the other side of their trades. They are also prone to trade in large blocks so as to maximize the value of their information.

Value Motivated

Value motivated traders use judgment based on careful research to identify the proper price at which to buy a stock. They will only trade if the price is at or below their perception of true value. They often use limit orders and patiently wait for the stock to reach their price. Trading can occur infrequently, but over time they may gradually accumulate or distribute large positions relative to their total portfolio.

Liquidity Motivated

Liquidity motivated traders transact because they need to. Perhaps a better opportunity has come along, or they may simply need to raise cash for some reason. As such, they are not particularly sensitive either to valuation or to unique information.

Since liquidity motivated traders are natural counterparties to value and information traders, their existence in the market has value. Liquidity traders can try to recognize the value they add by offering orders inside the spread as a way of attracting other traders to them.

Passive Traders

Index managers have to trade in order to rebalance their positions to the benchmark whenever there is a change in benchmark constituency. Sometimes (for example, if a company being added is larger than the one it replaces) this will require substantial liquidity, as the trade can have flow-through effects to many other stock weights in the index. Passive managers are also very sensitive to trading costs, as they want to limit their tracking error relative to a benchmark that does not incur such costs.

Passive traders often resemble dealers by providing the bid (or ask, as appropriate) on their stocks of interest and allowing the opposing party to decide when to trade. This allows them to exchange their lack of urgency for a lower execution cost.

Theory suggests that trading costs are related to various characteristics of the stock, the market and the trader. These include:

• Liquidity
• Risk
• Size of trade (relative to liquidity)
• Momentum
• Trading style

These relationships can be fed into a nonlinear regression model to estimate trading costs prior to executing the trade. This can be useful both in helping a fund manager decide the proper size of a given order and also to compare the actual trading costs to the original estimate as a measure of execution quality.

Volume-weighted average price (VWAP) is the average price at which a stock trades in a given period, weighted by the volume traded at each price. Many investors use the difference between their own execution price and the VWAP as a measure of their execution costs (specifically the market impact.)

The advantages of using VWAP to measure trading costs are that it:

• is easy to compute
• is easy to understand
• can be computed quickly, which helps traders make execution decisions in real time
• works best when measuring relatively small trades in markets without a strong trend

Disadvantages of VWAP include:

• the fact that it doesn’t account for costs related to delays or trade cancellation (implicit costs such as the stock “getting away”)
• it can be misleading if the trader’s volume is a significant percentage of the total volume
• it is not sensitive to the prevailing market conditions or the size of the trade, which both impact execution
• it can be gamed by delaying trades whenever the market price is above the current VWAP