Michael Dowdy Email & Phone Number

Washington, Dc, Us

Mclean, Va, Us

Mclean, Va, Us

Head of Digital Analytics for Market & Liquidity Risk, Counterparty Credit Risk Modeling-Led cross-functional product development for cloud-based digital analytics application to support model hosting, dashboarding, and data pipelines while eliminating dozens of manual processes and applications-Led team of three data scientists to build rare-events machine learning models that use natural language processing (NLP) to scan call transcripts for competitive information, an automation that reduced labor hours of multi-day manual processes by 90+% while producing 300+% greater effectiveness-Led team of three data scientists to build and manage suite of counterparty credit risk rating models and their limit setting framework, improving classification and variance (50% reduction in manual overrides) while saving hundreds of thousands in annual vendor costs-Designed custom text summarization and clustering algorithms for processing millions of documents and deployed those algorithms in a production application for hundreds of users

Worldwide, Oo

-In collaboration with cross-functional product team, invented predictive modeling design at the core of Deloitte's Solvas|ALLL+ CECL credit loss prediction software that was successfully launched at more than a dozen of the largest US banks, generating millions in annual revenue-Led international team of 10-12 data scientists and managers to design and build credit loss predictive modeling and time-series forecasting programs at seven of the largest US banks, managing more than $30MM in revenue-Led international team of 10-12 data scientists and managers to critically review and validate the predictive modeling programs at nearly a dozen of the largest US banks, managing more than $10MM in revenue

Worldwide, Oo

-Led team of 3-6 data scientists to design and build DFAST modeling programs with complete credit, pre-provision net revenue (PPNR), and ALLL stress testing model suites for several banks crossing the $10B DFAST regulatory asset threshold, including ‒Loan-level logit and general estimating equations MF/CRE models ‒Ratings-based parameterized transition matrix C&I models ‒Hazard SFR/retail models, such as Cox proportional hazards ‒Deposits, securities, and other PPNR dynamic regression time series models (i.e., ARIMAX) ‒Aggregate credit loss ARIMAX and panel AR time-series models

Worldwide, Oo

-Led two data scientist teams to assess and enhance credit, PPNR, and ALLL stress testing model suites for several banks approaching $50B in assets to prepare DFAST modeling programs for CCAR-Guided banks implementing PD/LGD ALLL models and dual risk rating systems, including full custom ALLL builds and a suite of customized CRE and C&I PD/LGD vendor scorecard models -Valued acquired loan portfolios for day one valuations and goodwill impairment with Markov transition matrix PD models and LGD hazard curves-Authored an econometric analysis that disentangles the long-term effects of specialty-drug pharmaceutical innovation on the quality of life and productivity of patients

Worldwide, Oo

-Designed rare-events analysis and sampling techniques to identify fraudulent/suspicious transactions, money laundering, and illicit drug distribution, across retail, banking and manufacturing industries-Invented a securities portfolio risk classification model by simulating hypothetical retirement portfolio annualized standard deviations of returns to measure the risk-return performance of broker dealers -Created process to link and analyze disparate survey data, regulatory filings, and web-scraped financials in multi-year market trend analysis supporting the Secretary of Labor’s annual reports to Congress

College Park, Md, Us

Thesis: “Growth, Institutions, and Colonial Heritage”Advisor: Professor John Joseph WallisAnalyzed the relationship between length of colonization and present day income per-capita across former colonies, employing various econometric techniques, such as two-stage least squares, and constructing an analysis dataset based on secondary research.

Master Of Science (M.S.), Applied Economics

Bachelor Of Arts (B.A.), Economics