With AlgoTrader any rule-based trading strategy can be automated, as the following real-world examples demonstrate
Medium- to long-term trend following (CTA)
Our client trades a standard yet very efficient example of this well-known group of systematic trading strategies. It trades a large number of underlyings, spanning virtually all asset classes, with necessary trading signals derived from daily price inputs. Given this strategy’s extensive use of future contracts, sophisticated roll mechanisms – specific to each asset – needed to be implemented.
Multi-indicator Forex trading
This client focuses on Forex spot trading and employs an intraday strategy based on a number of technical indicators. While some of these indicators are time-related, others may trigger a trade at any time during the session. The pure intraday character of the strategy requires all positions to be closed before market close. AlgoTrader ensures that this happens each and every day.
This model trades a large number of options, both listed and OTC, based on various underlyings. So it has to accommodate and aggregate different inputs, such as custom volatility surfaces. The building blocks of the strategy are combinations of instruments, such as strangles, butterflies and more complex, custom derivative spreads, which the system has to treat as one sole position. To do this AlgoTrader creates synthetic positions that aggregate all necessary information from their constituents.
Equity statistical arbitrage
This high-frequency trading model continuously looks for short-term price discrepancies in various stock markets around the globe. In order to seize these opportunities as soon as they emerge, very large volumes of data need to be processed and stored at ultra-low latencies. Once the trading signal is generated, intelligent order execution takes over to minimize potential slippage.
This client uses AlgoTrader to monitor the performance of a large number of historically correlated security pairs. When the correlation between two securities demonstrate a temporary weakness, a pairs trade is opened by shorting the outperforming stock and going long on the underperforming stock. The strategy uses the AlgoTrader – Pair Trading Lab integration to select candidate pairs from a database of more than 10 million pre-analyzed U.S. equity pairs.
This strategy is composed of several sub-strategies with varying complexity and trade frequency. While different modules may trade identical instruments (based on different logic), these modules are able to communicate with each other on a tick-by-tick basis without exceeding complex risk limits at the portfolio level.
Social Media Trading
This client has developed a proprietary algorithm, which allows it to mine the web and social media in “real time” for information that can be used to predict security pricing. For this, it uses big data processing, semantic analysis and machine learning to map the changes in the content of virtual conversations and relate these to changes in security pricing by training the algorithms using historic data. The client’s model can predict changes in securities pricing which are then traded using AlgoTrader.
This client engages in selective market making on number of exchanges and instruments by using a proprietary pricing model. An important part of the strategy implementation was the creating of sophisticated risk management system which allows monitoring of exposure levels over instrument type, currency, account and portfolio.
Macro / Special situation
Our client needed to rapidly implement a specific strategy to exploit a lucrative yet short-lived opportunity created by the ambient macro-economic conditions. The model traded both futures and options on various time-frames in combination with continuous hedges. The strategy was implemented successfully within a few days and delivered the expected results.