Automated option trading pdf


In effect, greater certainty comes at a significant opportunity cost of vastly reduced access to liquidity. Although the spread books do have some liquidity, in this fragmented high frequency world, finding the best possible execution in sufficient size requires a platform that can aggregate the largest possible number of liquidity sources for both equities and options. Moving from trading the "package" to managing aggregated liquidity across multiple venues introduces additional execution risks.

The right algorithmic execution tools are needed to negate "leg" risks. Optimal execution requires a single consolidated transparent interface and sophisticated execution algorithms. It also has the virtue of making the most efficient possible use of available desktop real estate. However, while existing pair algos may work well for equity pairs, a different approach is often required for equity options.

Liquidity in the options market can have a very different profile than equities; naively applying a conventional equity pair algo to the options market is susceptible to market impact by continually pushing to execute. For example, an algorithm that constantly trades when the pair hits the desired spread may hold the market from providing price improvement and potentially signal to the market that there is an algorithm at work. While sourcing equity and option liquidity across multiple venues has implicit legging risks, these can be minimised with the right tools.

Most applications have a rather limited response when a leg gets hung up; they either immediately go to market or pop up an alert.

Neither response is particularly helpful; the first increases execution costs perhaps substantially , while the second compels the trader to react possibly prematurely.

Better alternatives include allowing the trader to specify a time limit before further action, in case the market drifts back in the right direction of its own accord. If this doesn't happen, a second stage would be to enable the trader to grant the application a specific degree of price discretion.

A further method of maximising control of a multi-legged trade that includes equity options is pegging. In an ideal world, the trader needs the facility to peg each leg of a trade to a benchmark, which could be volatility or delta or just a target spread value.

Upper and lower bounds set on either side of these benchmarks can then be used to control the execution algorithm behaviour, such as pauses. A further advantage is if the pegging is in line with a market benchmark. As the underlying equity rallies, the trade becomes less attractive. Strategies that combine equities and options can easily create a risk management and book-keeping nightmare. Therefore any trading application needs to be tightly integrated with both the trader's blotter and middle and back office systems.

They offer traders "spread" books that allow quoting and trading in "pre-packaged" spreads. While this offers a measure of certainty i.

In effect, greater certainty comes at a significant opportunity cost of vastly reduced access to liquidity. Although the spread books do have some liquidity, in this fragmented high frequency world, finding the best possible execution in sufficient size requires a platform that can aggregate the largest possible number of liquidity sources for both equities and options. Moving from trading the "package" to managing aggregated liquidity across multiple venues introduces additional execution risks.

The right algorithmic execution tools are needed to negate "leg" risks. Optimal execution requires a single consolidated transparent interface and sophisticated execution algorithms. It also has the virtue of making the most efficient possible use of available desktop real estate.

However, while existing pair algos may work well for equity pairs, a different approach is often required for equity options. Liquidity in the options market can have a very different profile than equities; naively applying a conventional equity pair algo to the options market is susceptible to market impact by continually pushing to execute.

For example, an algorithm that constantly trades when the pair hits the desired spread may hold the market from providing price improvement and potentially signal to the market that there is an algorithm at work. While sourcing equity and option liquidity across multiple venues has implicit legging risks, these can be minimised with the right tools.

Most applications have a rather limited response when a leg gets hung up; they either immediately go to market or pop up an alert. Neither response is particularly helpful; the first increases execution costs perhaps substantially , while the second compels the trader to react possibly prematurely. Better alternatives include allowing the trader to specify a time limit before further action, in case the market drifts back in the right direction of its own accord.

If this doesn't happen, a second stage would be to enable the trader to grant the application a specific degree of price discretion. A further method of maximising control of a multi-legged trade that includes equity options is pegging.

In an ideal world, the trader needs the facility to peg each leg of a trade to a benchmark, which could be volatility or delta or just a target spread value. Upper and lower bounds set on either side of these benchmarks can then be used to control the execution algorithm behaviour, such as pauses. A further advantage is if the pegging is in line with a market benchmark. As the underlying equity rallies, the trade becomes less attractive.