Bio

Rebecca Knowlton is a PhD Candidate studying complex heterogeneity in the utility of surrogate markers under the supervision of Layla Parast. She passed her written preliminary exam in Spring 2021 and her oral candidacy exam in Spring 2023, and she expects to graduate in Spring 2025. During her doctoral program she has completed data science internships with Apple and Microsoft, applying statistical methodology to tackle industry problems like fraud prevention, application security, and enhanced buildout efficiency of cloud services. Prior to graduate school, Rebecca earned her Bachelor of Arts in Mathematics at Trinity University, Summa Cum Laude. She worked as an actuarial analyst for three years, pricing automobile insurance and passing the first five CAS exams, before returning to school to pursue her doctorate in Fall 2020.

Education

University of Texas at Austin | Austin, TX

Ph.D. in Statistics | Expected May 2025

Trinity University | San Antonio, TX

B.A. in Mathematics, Summa Cum Laude | May 2017

Research Experience

Department of Statistics and Data Sciences, University of Texas at Austin | Graduate Researcher | 2022 - 2025

  • Proposed novel measures for the proportion of the treatment effect on the primary outcome that is explained by the treatment effect on the surrogate (PTE) with respect to multiple baseline covariates, via parametric and semiparametric procedures and including a bootstrapped variance estimation
  • Developed a global test for evidence of heterogeneity, as well as a method for identifying particular subregions of strong surrogacy
  • In the case of heterogeneous surrogate utility, proposed a fully nonparametric method for efficient testing using surrogate information (ETSI), which enables treatment effect estimation and hypothesis testing in a setting where the surrogate is valid to substitute for the primary outcome for certain patient subgroups, and not for others
  • Demonstrated new methodology on simulated data in a variety of settings, and illustrated methods on AIDS clinical trial data.

SISTM Team, Bordeaux Population Health Center, University of Bordeaux | Visiting Graduate Researcher | November 2023

  • Presented dissertation research on complex heterogeneity in the utility of surrogate markers in department seminar and discussed future directions for the work based on feedback and overlapping interests with the SISTM team

Department of Information, Risk and Operations Management, University of Texas at Austin | Graduate Research Assistant | Fall 2022

  • Researched methods for developing AI to advise medical experts in high-stake decision-making, such as organ transplants
  • Developed methods aimed to complement human expertise and judgement via interpretable models while minimizing information leakage

Industry Experience

Microsoft | Data Science Intern | May 2024 - August 2024

  • Implemented a tool in C# to migrate existing Managed Service Identity (MSI) configurations to a modern, platform-agnostic configuration library that enables the combining of multiple runtime files into a single flattened file
  • Supported modernized feature management and zero-touch enablement of new environments in this migration, validating functional equivalency to old deprecated configurations

Microsoft | Data Science Intern | May 2023 - July 2023

  • Piloted metrics to monitor efficiency and drive improvements for the building out of new cloud environments for Microsoft Identity services
  • Served as the sole data scientist on a software engineering team and collaborated with the engineers to understand the complex dependencies of the buildout process and associated data

Microsoft | Data Science Intern | May 2022 - August 2022

  • Researched methods for evaluating application health within Azure Active Directory, including synthesizing information across eight datasets and understanding best practices for modern identity management
  • Created measure of application ``blast radius'' to capture how interconnected an application is within an organization in the event of a security incident

Apple | Data Science Intern | May 2021 - August 2021

  • Researched, engineered, and implemented new features for the primary fraud decisioning model actioning all purchases on the App Store
  • Collaborated cross functionally with four different teams to understand unexplored dataset of 9 billion rows, and trained machine learning model to assess potential feature lift

USAA | Actuarial Analyst | June 2017 - July 2020

  • Spearheaded pricing efforts for automobile insurance in Texas, the company's largest state at $1.9 billion written premium, and assisted in implementing behavior-based insurance and improved rate capping methodology
  • Led a team of five to analyze countrwide Auto loss trends and understand the drivers for frequency and severity, adjusting data as needed and making appropriate recommendations for annual Auto rate planning
  • Mentored new analysts through technical processes, including rating factor revisions and rate indications

Teaching Experience

Head Teaching Assistant | University of Texas at Austin | Spring 2024 - Fall 2024

  • DS 395T: Data Science for Health Discovery and Innovation

Course Co-Developer | University of Texas at Austin | Fall 2023

  • DS 395T: Data Science for Health Discovery and Innovation

Teaching Assistant | University of Texas at Austin | Fall 2021 - Spring 2022

  • SDS 320E: Elements of Statistics
  • SDS 324E: Elements of Regression Analysis

Tutor | Advantage Testing, Austin, TX | 2022 - Present

  • ACT/SAT
  • Mathematics
  • Statistics
  • Writing

Rebecca Knowlton


Bio

Rebecca Knowlton is a Statistics PhD student in the Department of Statistics and Data Sciences at the University of Texas at Austin. After earning her Bachelor of Arts in Mathematics at Trinity University, she worked as an actuarial analyst at USAA for three years while passing the first five exams with the Casualty Actuary Society. During this time, she served as the sole auto pricing analyst responsible for Texas, the company's largest state at roughly $1.9 billion written premium, and led the auto pricing team for loss trend analysis. She became interested in the other ways in which statistics and data science can be leveraged to solve complex problems, so she decided to return to school to begin a career in statistics research. Her research interests include machine learning, causal inference, and Bayesian statistics.

Education

University of Texas at Austin | Austin, TX

Ph.D. in Statistics | August 2020 - Present

Trinity University | San Antonio, TX

B.A. in Mathematics | May 2017

Experience

Microsoft (Upcoming) | Data Science Intern | May 2022 - August 2022

Apple | Data Science Intern | May 2021 - August 2021

  • Researched, engineered, and implemented new features for the primary fraud decisioning model actioning all purchases on the App Store
  • Collaborated cross functionally with four different teams to understand unexplored dataset of 9 billion rows, and trained machine learning model to assess potential feature lift

USAA | Actuarial Analyst | June 2017 - July 2020

  • Spearheaded pricing efforts for automobile insurance in Texas, the company's largest state at $1.9 billion written premium, and assisted in implementing behavior-based insurance and improved rate capping methodology
  • Led a team of five to analyze countrwide Auto loss trends and understand the drivers for frequency and severity, adjusting data as needed and making appropriate recommendations for annual Auto rate planning
  • Mentored new analysts through technical processes, including rating factor revisions and rate indications