Dynamic Matching Engine for Trade Execution

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current securities market systems, particularly Continuous Limit Order Book (CLOB) systems, face inefficiencies such as adverse selection and market impact, which disadvantage institutional investors by allowing short-term traders to exploit informational advantages and impose significant trading costs.

Innovation Solution

Implementing a machine learning engine to dynamically calibrate and control matching engine rule sets, optimizing matching times and order execution parameters based on real-time and historical market data to minimize adverse selection and market impact, while maximizing liquidity.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If a CLOB-based system is used to enable quick matching and speed of execution, then immediacy of matching is improved, but adverse selection and market impact increase for institutional investors

Engineering Contradiction:

Improvespeed of executionVSAvoidadverse selection

Core Design Contradiction:

SpeedVSObject-affected harmful factors

Solution Approach 1:

The patent implements dynamic matching intervals that adjust based on market conditions and order characteristics. Instead of continuous matching, the system dynamically determines matching intervals to balance execution speed with reducing adverse selection, allowing the system to adapt to changing market environments and investor needs.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system changes the matching parameter from continuous to interval-based, and further optimizes by dynamically adjusting the interval duration based on market volatility, order size, and other factors. This parameter change allows the system to maintain speed while reducing the harmful effects of adverse selection.

Inventive Principle:

Principle #35Parameter changes

2Quantity of substance

If a CLOB-based system is used to enable quick matching, then liquidity provision is improved, but trading costs for institutional investors increase

Engineering Contradiction:

ImproveliquidityVSAvoidtrading costs

Core Design Contradiction:

Quantity of substanceVSLoss of energy

Solution Approach 1:

The system dynamically adjusts matching intervals based on liquidity conditions and order characteristics. For large institutional orders, it extends matching intervals to reduce market impact and trading costs, while for smaller orders it maintains faster matching to preserve liquidity provision benefits.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system applies partial matching by not continuously matching all orders at all times. Instead, it selectively matches orders during optimized intervals, providing sufficient liquidity for market function while reducing excessive trading costs for institutional investors.

Inventive Principle:

Principle #16Partial or excessive action

3Speed

If matching occurs continuously in a CLOB system, then order execution speed is improved, but the ability of market participants to forecast order details worsens

Engineering Contradiction:

Improveorder execution speedVSAvoidinformational advantage

Core Design Contradiction:

SpeedVSLoss of information

Solution Approach 1:

The patent extracts the continuous matching mechanism and replaces it with interval-based matching. This removal of continuous matching eliminates the informational advantage that short-term traders gain from real-time order flow visibility, while still providing sufficient execution speed through optimized matching intervals.

Inventive Principle:

Principle #2Taking out (Extraction)

Data Source

PatentUS20210272201A1Systems for optimizing trade execution

Publication Date: 2021.09.02 GINIS ROMAN

AI summary

Systems and methods for optimizing trade execution by computing market reaction to recent trades of a security; calculating matching parameters for the security in response to the computed market reaction and at least one of historical market data and real-time market data; calculating a trade window for the next match; and executing the trade during the window.