Arman Oganisian

Arman Oganisian

Assistant Professor of Biostatistics

Brown University

I am currently an Assistant Professor of Biostaistics (tenure-track) at Brown University’s Department of Biostatistics. My methodological research centers around developing Bayesian methods for flexible causal effect estimation with observational data. These methods blend principled causal reasoning, nonparmetric Bayesian modeling, and efficient computation to build systems for data-driven decision making.

Most of my motivating applications have been in oncology - e.g. Bayesian nonparametric cost and cost/efficacy of estimation for endometrial cancer treatments; assessing treatment effect heterogeneity of photon therapy across cancer types. A related interest, that is inseparable from this work, is in methods for efficient MCMC computation. My current research is partially funded by a PCORI grant and focuses on developing Bayesian semiparametric methods for estimating and optimizing effects of sequential treatment strategies with applications in leukemia.

I received my PhD in Biostatistics from the University of Pennsylvania under the supervision of Jason Roy and Nandita Mitra. Here you will find a link to my CV and a selection of some past and current research, talks, etc. I sometimes blog about statistics, Bayesian methods, computation/MCMC, and other things I happen to stumble upon during research.

Interests
  • Bayesian nonparametrics
  • Causal Inference
  • Sequential decision-making
  • Health Economics
  • Oncology
Education
  • PhD in Biostatistics, 2021

    University of Pennsylvania

  • MS in Biostatistics, 2018

    University of Pennsylvania

  • BA in Quantitative Economics, 2013

    Providence College

Projects

*
Collaborative Causal and Bayesian Modeling

Collaborative Causal and Bayesian Modeling

Collaborative data analysis projects using Bayesian and/or Causal methods.

Non-parametric Bayes

Non-parametric Bayes

Bayesian modeling - flexiblilty, uncertainty quantification, full posterior inference.

Bayesian Causal Inference

Bayesian Causal Inference

Shrinkage, partial pooling, nonparametrics, and sensitivity analysis via priors - just some of the value Bayesian modeling can add to causal inference.

ChiRP

ChiRP

R Package for Dirichlet Process Mixtures of zero-inflated, logistic, and linear regressions.