Quantitative Modeling Lead [Multiple Positions Available]

Duties: Design and monitor model performance metrics.

Review and build benchmark model for retail credit risk related models.

Assess and help mitigate the model risk of complex models used in the context of credit card scorecard models for decision-making purposes.

Validate models by evaluating conceptual soundness.

Design and conduct experiments to compare a model's prediction against actual outcomes or against the output of alternative benchmark models.

Own validation projects, including planning, execution, coordination, and final delivery.

Drive all phases of the model review process while also contributing individual technical assessments.

Conduct validation of regulatory models related to capital stress testing.

Drive coordination of the Annual Status Assessment (ASA) process for the Consumer and Community Banking line of business.

Conduct no-change re-reviews and minor model enhancement reviews.

QUALIFICATIONS:

Minimum education and experience required: Ph.D.

in Aeronautics and Astronautics, Engineering (any), Statistics, Computer Science, Mathematics or related field of study plus one (1) year of experience in the job offered or as Quantitative Modeling Lead, Research Assistant, or related occupation.

The employer will alternatively accept a Master's in Aeronautics and Astronautics, Engineering (any), Statistics, Computer Science, Mathematics or related field of study plus three (3) years of experience in the job offered or as Quantitative Modeling Lead, Research Assistant, or related occupation.

Skills Required: This position requires experience with the following: conducting model development, risk management, underwriting, or portfolio performance analysis for credit card portfolios, mortgage products, or auto services loans within the financial services industry; building, validating, and interpreting logistic regression and Linear Regressions models using real-world data that is collected from actual events, processes, or observations in daily situations rather than generated from idealized, simulated environments; assessing projects where models were used for classification and regression tasks, evaluating the model's metrics, and assessing how the results were applied to drive business decisions; implementing machine learning solutions end-to-end, including data preprocessing, algorithm selection, model training, and deployment in a production or research setting; time series modeling, including trend, seasonality, and forecasting analyses; programming in R or Python to perform data analysis, model development, or automation of analytical processes; formulating hypotheses, selecting modeling techniques, back-testing predictions, and documenting performance metrics to ensure robustness and regulatory compliance; applying Xgboost to improve model predictive performance; applying hyperparameter optimization techniques and applying explainability methods to assess feature importance, including using Shapley values for model inter...