Paychex: Risk Modeling Analyst II

Perform high level data mining and analytical projects, develop predictive models, understands specific business details, and interact with data owners/key stakeholders.

Paychex Company: Paychex
Location: Rochester, NY

Candidates can apply directly to:

About Paychex
With 100+ office locations across the nation, Paychex produced $2.3 billion in revenue in fiscal 2013. The company offers an ever-growing variety of payroll and human resource products and services that help clients do what they do best - run their business. With a wide range of services, including payroll processing, retirement services, insurance, and a fully outsourced human resource solution, Paychex customizes its offering to the client's business, whether it is small or large, simple or complex.

Paychex is always looking for individuals who want to work for a company that allows for growth and development. With your desire to succeed and our training and resources, opportunities at Paychex are more than just jobs. We provide an atmosphere that fosters a healthy work life balance, and our comprehensive benefits package provides health care, retirement planning, education assistance, and much more.

Job Description:

Performs high level data mining and analytical projects for the development of predictive models. Understands specific business details related to the necessary data and/or conducts interviews with data owners/key stakeholders as necessary to clarify business data needs. Researches new theories and technologies in the predictive analytics world and finds applications within Paychex. Assists with development of results reporting.
  • Develops and implements scorecards/data driven predictive models covering different areas of the company, including Enterprise Risk Management, Marketing, Sales and Field Operations.
  • Utilizes advanced functionality of Microsoft Office SQL, VBA, SAS and/or Business Objects for data mining and automation. Extracts and houses data from internal Paychex systems for use in modeling applications.
  • Explores, validates and maintains high quality data with optimal data storage design. The data is used for model development and reporting.
  • Explores and gathers external data including both public and private sectors to complement internal company data.
  • Researches new theories and technologies in predictive analytics, generates ideas for transforming data into practical models, and explores applied research ideas and formulation of R&D projects to be applied to any aspect of the company.
  • Conducts analysis to validate the performance of existing models and recommends changes for improvements.
  • Develops custom systems designed to determine and/or track the effectiveness of new and existing models.
  • Executes all phases of quantitative research projects by identifying the business problem, conducting data exploration, modeling and communication of final results to Enterprise Risk Management leadership and other interested parties.
  • Shares knowledge within the Modeling and Analytics Group for cross training and development.
  • During model development works with business partners to understand all aspects of the business, from understanding business issues/priorities to analytically solving these business issues/priorities.
  • Provides consultation to all company users in the areas of risk model development, account monitoring techniques, credit process strategies for manual and automated underwriting, quantitative financial analysis and/or potential risk management programs for new ventures.

  • Bachelor's degree in IT, Statistics, Mathematics, Physics or Engineering, particularly Masters or PhD is preferred.
  • A minimum of 5 years of experience is required in any of the following subjects: data mining, data warehousing, regression analysis, predictive modeling.
  • Expert level proficiency in MS Excel and MS Access is required. Experience with SAS or MATLAB required. Experience with SQL, VBA or Business Objects is a strong plus. Solid understanding of regression and decision tree based methods for fitting predictive models is a strong plus. Ability to communicate in a concise and clear manner is required.