Liberal Arts and Sciences
Ohnishi Lab
Led by Yuki Ohnishi · Statistics
What we study
The lab develops statistical methods for causal inference, focusing on Bayesian approaches and complex data structures like randomized trials. They also study differential privacy and privacy-utility trade-offs in statistical modeling.
What working here is like
- Computational
- Reading, writing & theory
- No organisms (data only)
- Coding required
- Papers
- 4
Recent papers, explained
- Calibrated Bayes Analysis of Cluster‐Randomized Trials
2026 · Statistics in Medicine
Develops a Bayesian method to analyze cluster-randomized trials, focusing on treatment effects and cluster size.
- A Bayesian Nonparametric Approach to Mediation and Spillover Effects with Multiple Mediators in Cluster-Randomized Trials
2025 · Journal of the American Statistical Association
Creates a Bayesian approach to study mediation and spillover effects in cluster trials with multiple mediators.
Show 2 more papersShow fewer
- Inferring Causal Effect of a Digital Communication Strategy under a Latent Sequential Ignorability Assumption and Treatment Noncompliance
2024 · Journal of the American Statistical Association
Uses Bayesian methods to assess digital communication strategies, accounting for targeting rules and noncompliance.
- A Bayesian Analysis of Two-Stage Randomized Experiments in the Presence of Interference, Treatment Nonadherence, and Missing Outcomes
2022 · Bayesian Analysis
Proposes a Bayesian method for two-stage randomized experiments, addressing interference, nonadherence, and missing data.
Plain-language summaries are AI-written and reviewed by us; check the paper for details.
How to reach out
Yuki Ohnishi hasn't set up this page yet. You can also look them up in the UF directory (search “Yuki Ohnishi”).