Assumed risk versus actual risk: The new world of behavior-based risk modeling
Level
Over the last few years, financial institutions have dedicated significant time and effort to identifying how data science and artificial intelligence can help them better understand customer behavior, combat financial crimes, and regulate anti-money laundering. However, the traditional “assumed risk” models they’ve been deploying are only leveraging KYC (know your customer) data and data that customers voluntarily disclose about themselves. While still valuable, “assumed risk” models only reveal the tip of the iceberg. What financial institutions need to adopt is an “actual risk” model—a data analysis model that considers customers’ transactional data and real-time behavior that can be used to track financial crimes and predict future behavior.
Viridiana Lourdes explains how AI/ML professionals and data scientists at financial institutions can adopt and integrate actual risk models with existing systems to enhance the performance and operational efficiency of the financial crimes organization. Join in to learn how actual risk models can reduce segmentation noise, utilize unlabeled transactional data, and spot unusual behavior more effectively.

Viridiana Lourdes
Ayasdi
Viridiana Lourdes is data scientist at Ayasdi with over 15 years of experience applying advanced statistical methods to challenging problems and implementing solutions through the use of latest technology. She’s developed statistical models for bid data based on the integration of quantitative research with advanced scientific computation and collaborative interdisciplinary applications in many fields, and she has practical experience in econometrics, portfolio management, asset allocation, portfolio construction, optimization, and risk management. Her computer programming experience includes professional-level applications with Python and statistical software R. She earned her doctor of philosophy and MS in statistics and decision sciences from Duke University and an MA in finance and a BA in actuarial sciences from ITAM. Her quantitative skills include Bayesian dynamic linear models, econometric models, time series and forecasting, predictive modeling, nonlinear models, generalized linear models, multivariate analysis, multifactor models, and classification trees.
Presented by
Elite Sponsors
Strategic Sponsors
Zettabyte Sponsors
Contributing Sponsors
Exabyte Sponsors
Content Sponsor
Impact Sponsors
Supporting Sponsor
Non Profit
Contact us
confreg@oreilly.com
For conference registration information and customer service
partners@oreilly.com
For more information on community discounts and trade opportunities with O’Reilly conferences
strataconf@oreilly.com
For information on exhibiting or sponsoring a conference
pr@oreilly.com
For media/analyst press inquires