Co-Trading Changepoint Detection for Market Manipulation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional techniques for detecting market manipulation in securities trading suffer from high false-positive rates due to difficulties in distinguishing artificial changes from natural changes, particularly in low-priced or illiquid securities, leading to inefficient identification of fraudulent activities.
Innovation Solution
A server-based system performs co-trading changepoint detection by analyzing transaction time series data across multiple securities, using filtering parameters, time series parameters, and changepoint analysis to identify abrupt disruptions indicative of market manipulation, generating potential fraud alerts and reducing false positives.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional techniques analyze transactions of individual securities to detect market manipulation, then detection capability is provided, but false-positive rate increases due to inability to distinguish artificial changes from natural changes
Solution Approach 1:
The patent combines analysis of multiple co-traded securities into a unified detection framework. By merging transaction data from several related securities and analyzing them collectively using changepoint detection, the system distinguishes coordinated artificial manipulation patterns from individual natural price fluctuations, thereby reducing false positives while maintaining detection accuracy.
2Difficulty of detecting and measuring
If conventional techniques detect spikes in trading activity or price for individual securities, then potential manipulation is identified, but difficulty arises in distinguishing artificial changes from natural changes such as earnings-related news or regulatory announcements
Solution Approach 1:
The patent segments the analysis by grouping securities into co-trade relationships and applying changepoint detection to each segment. This segmentation allows the system to isolate manipulation patterns specific to coordinated securities while filtering out general market noise such as earnings announcements or regulatory news that affect individual securities independently.
Solution Approach 2:
The patent introduces changepoint detection analysis as an intermediary mechanism between raw transaction data and fraud identification. This intermediary statistically identifies abrupt disruptions in co-trading patterns, serving as a filter that distinguishes artificial manipulation from natural market changes caused by external factors like news or regulatory announcements.
3Productivity
If analysis focuses on single-point outliers in transactions, price, or volume of individual securities, then potential fraud can be detected, but efficiency decreases due to high false-positive rates
Solution Approach 1:
The patent merges outlier detection across multiple co-traded securities rather than analyzing each security independently. By combining transaction, price, and volume data from related securities and applying changepoint detection, the system identifies coordinated outliers that indicate manipulation while filtering out individual securities' natural variations, thereby improving both efficiency and reliability.
Data Source
AI summary
A system, apparatus, method, and non-transitory computer readable medium for performing co-trading changepoint detection may include a server caused to receive a first raw dataset, the first raw dataset including a plurality of transactions for analysis, each transaction of the plurality of transactions associated with a user account of a plurality of user accounts, generate at least one transaction time series based on the first raw dataset, determine changepoints in the first raw dataset by performing changepoint detection analysis on the generated at least one transaction time series, and generate at least one potential fraud alert based on the determined changepoints.