Match Event Outcome Detection for Extreme Trading Value Changes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Electronic trading systems face challenges in rapidly detecting and responding to extreme market changes that do not reflect a true consensus of market participants' values, leading to unfair market conditions and potential financial losses due to high-speed trading and irrational trader behavior.
Innovation Solution
Implementing a mechanism to detect and respond to extreme value changes by scanning for rapid market fluctuations, pausing the market, setting trade price limitations, or notifying the operator, thereby mitigating the effects of extreme market moves.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Stability of the object's composition
If velocity logic detection systems are used to prevent undesirable changes in values, then market stability is improved, but detection completeness deteriorates as these systems cannot detect all undesirable situations
Solution Approach 1:
The detection system is segmented into multiple independent detection mechanisms: velocity logic detection for rapid value changes, stop logic detection for gap detection, and the new match event outcome detection for verifying trade fairness. Each segment handles specific detection tasks, collectively achieving comprehensive market monitoring while maintaining system stability.
Solution Approach 2:
The match event outcome detection is nested within the existing velocity logic framework. The system first uses velocity logic to detect rapid value changes, then applies match event outcome detection as a deeper verification layer to analyze whether individual trades contributed to the abnormal movement, creating a multi-layered detection structure that enhances completeness without sacrificing speed.
2Measurement precision
If stop logic detection systems are used to prevent undesirable gaps between reference and received values, then value accuracy is improved, but detection completeness deteriorates as these systems may not detect all undesirable situations
Solution Approach 1:
The system merges stop logic detection with match event outcome detection. Stop logic continues to monitor value gaps and accuracy, while match event detection analyzes the fairness and legitimacy of individual trades. Together, these combined mechanisms provide both precise value monitoring and comprehensive trade legitimacy verification, achieving complete detection coverage.
3Reliability
If the system pauses the market to mitigate extreme moves, then market fairness is improved, but productivity deteriorates due to trading interruptions
Solution Approach 1:
Instead of pausing the entire market, the system applies partial action by targeting only the specific trades or instruments involved in the abnormal movement. The match event detection identifies problematic trades individually, and mitigation is applied selectively to those cases, maintaining overall market productivity while ensuring fairness in affected areas.
Solution Approach 2:
The system introduces an intermediary analysis layer that examines match events between trades before applying mitigation. This intermediary detection mechanism identifies unfair trading patterns and isolates them for targeted intervention, allowing the rest of the market to continue operating normally while addressing fairness issues in specific cases.
4Speed
If the system monitors all match events in real-time, then detection speed is improved, but device complexity increases
Solution Approach 1:
The system performs preliminary filtering by establishing criteria for abnormal match events before real-time monitoring begins. Velocity thresholds, trade size limits, and pattern recognition rules are pre-configured, allowing the system to quickly identify and flag suspicious events without analyzing every single trade in detail, thus maintaining detection speed while reducing processing complexity.
Data Source
AI summary
A processing module may be centralized and coupled to multiple inputs from customers, and behave deterministically, e.g., programmed to depend on state, inputs and outputs. The rapid speed of automated trading systems implementing such a centralized, deterministic module, where all users can access a central limit order object, can quickly result in an object state that does not reflect a true consensus or desirable state. Accordingly the resulting problem is a problem arising in computer systems due in part to the high speeds of computer systems. The solutions disclosed herein are, in one embodiment, implemented as automatic responses and actions by a computing system.