SUMMARYThis work presents the Intelligent Trading Architecture (ITA), which is a new automated trading system architecture that supports multiple strategies for multiple market conditions through hierarchical trading signals generation. The central idea of the proposed system architecture is decomposing the trading problem into a set of tasks that are handled by distributed autonomous agents under a minimal central coordination. With this kind of architecture, we can take advantage of currently available and future high-performance computing systems. These systems, due to the way computer architecture has evolved in the recent past and foreseeable future, are composed of multiple processor cores. We are implementing the ITA software architecture employing the Carnegie Mellon Navigation (CARMEN) robot control software and using a publish/subscribe communication model. Together, CARMEN and this communication model allow the implementation of high-performance, scalable parallel computing systems that leverage the architecture of multi-core systems. For this work, we evaluated the data structures and algorithms employed by the symbol module of the ITA software architecture, which is responsible for maintaining the synchronized local copies of exchanges limit order books (LOB) for the instruments traded by the system. Our LOB implementation strongly outperformed a reference implementation in all evaluated parameters by more than one order of magnitude in some cases, achieving average throughputs of 4 million orders/s when creating new orders, 3 million orders/s when changing existing orders, and 17 million orders/s when querying orders. Copyright
This work presents a new weightless neural network-based time series predictor that uses Virtual Generalized Random Access Memory weightless neural network to predict future stock returns. This new predictor was evaluated in predicting future weekly returns of 46 stocks from the Brazilian stock market. Our results showed that Virtual Generalized Random Access Memory weightless neural network predictors can produce predictions of future stock returns with the same error levels and properties of baseline autoregressive neural network predictors, however, running 5,000 times faster.
This work proposes a new automated trading system (ATS) architecture that supports multiple strategies for multiple market conditions through hierarchical trading signals generation employing h-signals, which are trading signals that are generated using other trading signals. The central idea of the proposed system architecture is to decompose the trading problem into a set of tasks handled by distributed autonomous agents under a minimal central coordination. We implemented the proposed ATS using a software architecture that employed a publish/subscribe communication model. In the current stage of development, we are able to run our ATS in back-test mode with moving-average crossover strategies on minute-by-minute market databases. We achieved very satisfactory performance results, processing 306.791 database rows representing more than two years of data in only 47 seconds.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.