Samuel Paul Gyamfi, PhD
Samuel Paul Gyamfi, PhD
I’m a Data Scientist at State Farm, working within P&C Actuarial & Underwriting Modeling, where I apply statistical modeling, machine learning, and quantitative methods to insurance risk, pricing, underwriting, and decision support.
I hold a PhD in Systems Modeling & Analysis from Virginia Commonwealth University (VCU), an MS in Applied Mathematics from Ohio University, a BSc in Actuarial Science from KNUST, Ghana, and an Advanced Diploma in Insurance from the Chartered Insurance Institute, UK. This multidisciplinary background allows me to work at the intersection of data science, actuarial analytics, mathematics, and insurance.
My technical toolkit includes Python, R, and SQL, with experience across generalized linear models, gradient boosting, survival analysis, ARIMA/GARCH models, Bayesian inference, Monte Carlo simulation, optimization, forecasting, and predictive modeling. I enjoy developing models that are not only technically rigorous, but also interpretable, reproducible, and useful for real-world decision-making.
Before joining State Farm, I worked as a Reinsurance Analyst at Phoenix Insurance, supporting treaty and facultative placements, developing loss and exposure analyses, contributing to pricing indications, and collaborating with brokers and cedents on coverage and claims development. I also served as a Branch Manager at Quality Life Assurance, where I led underwriting and customer operations, managed performance and sales targets, and worked closely with actuarial and claims teams on product and pricing decisions.
My work is centered on turning complex data and mathematical ideas into practical solutions. I particularly enjoy taking models from research and experimentation through validation and implementation—using backtesting, scenario analysis, and model monitoring to assess performance, while communicating results clearly to technical and non-technical stakeholders.
At the core of my work is a simple goal: use data, mathematics, and thoughtful modeling to make better decisions in complex and uncertain environments.