THE AML CHALLENGE - AND THE Kx SOLUTION
AML teams have the onerous task of validating the probity of their customers, the transactions they execute and, in many cases, the actions the institution undertakes on their behalf. As a result, they are striving to achieve the same levels of data aggregation and analysis that their trading counterparts have had to meet for MAR, MiFID and similar regulations. In the process they are facing the same problems in the sourcing, cleansing, analysis and governance of the required data, much of which is often siloed in different systems, stored in multiple formats and labeled with inconsistent nomenclature.
Kx for AML is designed to address precisely these challenges and provide real-time detection of prohibited activity ranging from individual trading misconduct and suspicious transaction profiles though to cross-party collusion and insider trading. The solution provides for both real-time and historical processing of bank transaction and trade data through a number of sophisticated algorithms, producing AML-specific alerts and case reports for compliance and investigation teams. The alerts range from the detection of fraudulent activity such as structuring of funds to transaction anomalies and activity within dormant accounts.
THE Kx SOLUTION
Kx for AML is based on the same market-leading technology used by regulators and brokers worldwide for trade and market surveillance. It is powered by the world's fastest time-series database, kdb+, for ingesting and normalizing data from multiple sources, performing real-time analytics for detecting AML violations, providing instant alerting and full lifecycle case management. The entire process is supported by a flexible HTML5-based UI providing investigative dashboards, market reconstruction and replay, flexible alert configuration and full backtesting capabilities
Click on the link below to read more about Kx for Surveillance
AML ALERTS AND REPORTS
Out-of-the-box alerts include:
- Flow Through of Funds - looks back through a user-defined number of days to determine if transactions in and out of a customer’s account exceed predefined thresholds
- Structuring - monitors transfers in or out of accounts, taking into consideration amounts being transferred and number of occurrences within a given period of time.
- Historical Deviation - checks if the amount of funds deposited or withdrawn deviates more than a certain percentage from the average of all deposits or withdrawals.
Other examples include checking on dormant accounts, risk counties and excessive transfers based on value and net worth. Alerts can be applied to the institution's customers or to its employees acting on behalf of those customers.
ALERTS MANAGEMENT AND TRACKING
All alerts generated by Kx for AML are posted to the AML Action Tracker along with information such as the Client ID of the entity, what stage of the workflow it is at and the alert ID. If a client has multiple alerts for different transactions they can be grouped together for easier investigation.
Alert details can be viewed in an accompanying side panel that can be expanded to further investigate the alert as well as progressing the ticket to different stages of a user-defined workflow e.g. "Under investigation", "Closed", "Reassigned". It also enables the user to directly query transaction data and reference data relevant to the alert.
CALIBRATION AND BACKTESTING
Backtesting enables organisations to test new alerts, fine-tune existing entries and run ad hoc analyses to retrospectively search for patterns of abuse not previously considered. Users can select a set of alerts and a data set against which to run. Once the alert logic has been executed it will display the results in a data grid for easy inspection. For each alert generated a ticket with the specific alert details is published to the action tracker where it can be further investigated
A configuration management screen allows users to set different thresholds for each alert in both backtesting and production environments. In addition to alert thresholds, users can configure the length of lookback periods over which analyses are run as well as filters on what data should be used.
DATA INGESTION AND MANAGEMENT
Transaction and reference data can be captured in both streaming and batch-delivered formats. Integrity checks ensure mandatory data fields are present with optional functionality for real-time transformation governed by a curated data dictionary.
The default schema covers transaction data, client reference data, trading summary data and rates data and can be easily extended to additional tables and sources. A series of secondary checks are carried out to determine and maintain the integrity of the data.
Once ingested the data can be retrieved via a variety of API calls designed to simplify and streamline the process of gathering data for detecting suspicious transactions and account activity
The basis for Kx Technology is a unique integrated platform which includes a high-performance historical time-series columnar database called kdb+, an in-memory compute engine, and a real-time streaming processor all unified with an expressive query and programming language called q.
Designed from the start for extreme scale, and running on industry standard servers, the kdb+ database has been proven to solve complex problems faster than any of its competitors.LEARN MORE