Trevor L. Wrobleski
| Home | Research | Projects & Awards |
Welcome
I am a Statistics PhD candidate at the London School of Economics, interested in using Bayesian methods, machine learning, and computational tools to solve complex problems in risk assessment, toxicology, and healthcare analytics. My advisors are Dr. Sara Geneletti and Dr. Francesca Panero. My research focuses on developing robust statistical models to improve decision-making in health, pharmacology/biotech, toxicology, and risk systems under uncertainty.
Research Interests
Statistical Methodology & Theory
- Bayesian Model Averaging & Stacking: Predictive distribution combination in M-open settings, scoring rules (log-score, CRPS), and coherency vs. predictive accuracy trade-offs.
- Optimal Experimental Design: Locally D-optimal and c-optimal designs for non-linear models, resource allocation strategies ($N$ vs. $G$), and design robustness under overdispersion.
- Robust Inference: Handling model misspecification, variance propagation in poly-$k$ survival adjustments, and quantifying bias in finite-sample BMD estimation.
Quantitative Risk Assessment
- Dose-Response Modeling: Hierarchical modeling of heterogeneity (chemicals/studies), benchmark dose (BMD) estimation, and derivation of empirical priors for regulatory science.
- Toxicology & Carcinogenicity: Analyzing large-scale bioassay databases (CPDB), accounting for litter effects and differential mortality, and improving points of departure (PoD) for FDA/EMA submissions.
Application Areas
Academic Research: Advancing Bayesian model averaging theory, developing robust methods for small-sample inference, creating open-source tools for reproducible risk assessment
Investment Decision Support: Quantitative due diligence for biotech assets, probability of technical success (PoTS) modeling, risk-adjusted NPV analysis for drug development pipelines, portfolio optimization under uncertainty
Regulatory Science & Strategy: Benchmark dose estimation for FDA/EMA submissions, prior sensitivity analysis for regulatory compliance, optimal study design to maximize approval probability while minimizing cost
Drug Development Optimization: Adaptive trial design, go/no-go decision frameworks, cross-regional data harmonization (US/EU/Asia), in-silico patient population generation for trial planning
Healthcare Data Science: Causal inference from real-world evidence, synthetic control methods for post-market surveillance, generative models for rare disease data augmentation
Education
- PhD Candidate, Statistics – London School of Economics
- MSc, Statistics (Research) – London School of Economics
- MMSc, Global Affairs, Economics and Business – Tsinghua University, Schwarzman College
- BA, Public Health – Johns Hopkins University
Skills
| Category | Proficiencies |
|---|---|
| Statistical Methods | Bayesian inference, Optimal experimental design, Hierarchical (mixed-effects) modeling, Survival analysis, MCMC |
| Programming | R (Tidyverse, Shiny), Python (NumPy, Pandas), SQL, Git |
| Bayesian & ML | Stan, PyMC3, TensorFlow, PyTorch, Scikit-learn |
| Optimization | Gurobi, CPLEX, PuLP |
| Languages | English (Native), French (Intermediate), Chinese (Mandarin/普通话) |
About Me
Outside of research, I enjoy running, tennis, hiking, reading, and playing piano.
You can contact me from my university page.