Brandon P. Pipher
Brandon P. Pipher

Statistician/Mathematician/Data Scientist

About Me

Mathematical Statistician at the U.S. Census Bureau.

Download CV
Education
  • MS in Applied Mathematics

    Kent State University

  • BS in Mathematics

    University of Akron

Experience

  1. Supervisory Mathematical Statistician

    United States Census Bureau

    Responsibilities include:

    • Lead data-driven projects leveraging advanced statistical and machine learning methodologies within the Decennial Statistical Studies Division’s Sampling Branch.
    • Spearhead research and development for the 2030 Post-Enumeration Survey (PES), improving coverage estimation by integrating administrative records with Census data and employing innovative modeling approaches.
    • Conduct research for the Continuous Count Study, enhancing intercensal population estimates through linkage of Census products, commercial data, and government administrative records. Applied statistical learning methods including Log-Linear and Latent Class modeling. Presented findings at the 2024 Joint Statistical Meetings and the 2024 Federal Committee on Statistical Methodology.
    • Designed and executed statistical programming for the 2020 Post-Enumeration Survey (PES), developing an Inmover probability imputation model and applying advanced feature selection to improve the accuracy of coverage estimates.
    • Developed and applied graph theory and network-based methods to enhance residence assignment and household inference from administrative data, improving accuracy of person-to-household linkage and residence imputation models.
  2. Quantitative Analyst

    Nations Lending

    Responsibilities include:

    • Partnered with Risk Management, Compliance, and Product teams to create automated reports and dashboards, providing insights on Key Performance Indicators (KPIs) and Objectives and Key Results (OKRs) using statistical modeling and data science techniques.
    • Delivered high-impact analytical summaries to senior leadership, developing flexible reporting solutions to drive strategic decision-making and monitor performance indicators.
    • Built time series forecasting models using public data to predict quarterly mortgage loan origination volume, optimizing workforce allocation and reducing operational costs.
    • Applied Natural Language Processing (NLP) to analyze mortgage process documentation, uncovering bottlenecks and reducing closing times through machine learning-based workflow improvements.

Education

  1. MS in Applied Mathematics

    Kent State University

    My studies included measure-theoretic probability and statistical computing.

    My research was on regression methods to induce sparsity, with a focus on non-convex methodologies.

    Read Thesis
  2. BS in Mathematics

    University of Akron

    My studies included topics in real analysis and abstract algebra.

    • Member of Phi Sigma Alpha: Buchtel College of Arts and Sciences Scholastic Honorary Society
    • Member and Treasurer of Pi Mu Epsilon: Mathematics Honorary Society (Ohio Nu Chapter)