Expoint - all jobs in one place

The point where experts and best companies meet

Limitless High-tech career opportunities - Expoint

JPMorgan CCB Risk Program Associate - Quantitative Modeler 
United States, Ohio 
950828813

17.08.2024

Job Responsibilities -

  • Design, develop, test, and validate statistical/economic models for consumer/retail portfolios, including probability of default, loss given default, and exposure at default.
  • Utilize state-of-the-art modeling including both classical statistical modeling approaches and modern machine learning approaches to enhance existing models and tackle challenging modeling problems
  • Manage end-to-end model development process, including data manipulation, exploratory data analysis and pattern discovery, model development, refinement and validation, documentation, assisting with implementation, and performance monitoring
  • Collaborate with cross functional partners in Risk, Finance, Technology, Model Governance throughout the entire modeling life cycle.

Required qualifications, capabilities and skills -

  • Advanced degree in a quantitative discipline (e.g. Mathematics, Statistics, Economics, Computer Science, Operations Research) - Masters with 2+ years of relevant working experience or a PhD.
  • Strong data analysis and statistical/economic modeling experience, such as generalized linear models, multivariate analysis and time series analysis
  • Proficiency in advanced analytical languages (e.g. SAS, Python, R);
  • Ability to work with large data and perform extensive analysis to draw useful insights
  • Strong communication skills to present to and collaborate with business partners and model end-users Strong organizational and multi-tasking skills with demonstrated ability to manage expectations and deliver quality results on time
  • Comfortable working both independently and in a team environment.

Preferred qualifications, capabilities and skills

  • Credit risk modeling experience is a plus, but not necessary.
  • Familiarity with framework of machine learning pipeline (e.g. tensor flow, scikit-learn) is not required but a plus.