Market-making strategy engine system based on low-delay complex events

Abstract

The invention discloses a market-making strategy engine system based on low-delay complex events, which greatly improves the convenience of strategy writing while ensuring running speed, and provides a solution for contradiction between speed and usability. According to the technical scheme, a CEP engine based on an LLVM is used for replacing a traditional C + + engine, the CEP engine provides a language interface easy to learn, meanwhile, in combination with the sequence, branch and loop structure oriented to a process language and the flexible and changeable characteristics of an SQL-like language, the CEP engine can quickly respond to the change of a rule (namely a quantitative strategy), and the secondary development and daily maintenance cost is reduced; and meanwhile, microsecond-level complex stream data processing is realized through LLVM, so that the requirements of low delay and high flexibility of stream data processing in the financial futures industry are comprehensively met. The invention further provides a user-defined domain-specific language DSL for strategy writing.

Description

technical field [0001] The invention relates to a market-making strategy engine system of a financial market maker, in particular to a market-making strategy engine system with low-latency characteristics realized based on low-latency complex event processing technology. Background technique [0002] Complex Event Processing (Complex Event Processing, hereinafter referred to as CEP) is an analysis technology based on event flow in a dynamic environment. Events here are usually meaningful state changes. By analyzing the relationship between events, using filtering, correlation, aggregation, etc. Technology, formulate detection rules based on the timing relationship and aggregation relationship between events, continuously query the event sequence that meets the requirements from the event stream, and finally analyze and obtain more complex composite events. Suitable scenarios for CEP include real-time risk management, real-time transaction analysis, cyber fraud, cyber attack,...

AI Technical Summary

Problems solved by technology

[0008] Most of these engines provide C++ interfaces for writing quantitative strategies, but due to the difficulty of C++ itself, the learning curve, and the difficulty of debugging, the writing of such strategies often requires professional IT personnel. For market makers who are mainly engaged in trading business Obviously, this is not a small burden

Images