Advanced Discrete-Event Simulations in R

28 June 2026, 8:35 AM – 12:00 PM (local time)

Faculty

  • Thomas A. Trikalinos, MD – Professor of Health Services, Policy & Practice and Biostatistics, Brown University School of Public Health, United States
  • Fernando Alarid-Escudero, PhD – Assistant Professor of Health Policy, Stanford University School of Medicine, United States
  • Yuliia Sereda, PhD – Researcher, Brown University, United States
  • Stavroula A. Chrysanthopoulou, PhD – Assistant Professor of Biostatistics, Brown University, United States

Course Overview

Advanced Discrete-Event Simulations in R is an advanced, hands-on workshop that focuses on the design and implementation of discrete-event simulation (DES) models using nonhomogeneous Poisson point processes (NHPPPs). The course abstracts DES to the core task of sampling time-varying event processes, emphasizing exact, efficient simulation algorithms and practical coding strategies in R.

Using guided exercises with the nhppp R package, participants will practice simulating complex event histories, including unconditional and conditional sampling, while prioritizing intuition, computational efficiency, and reproducible implementation for health decision modeling.

Learning Objectives

Participants will learn how to:

  • Understand key theoretical properties of nonhomogeneous Poisson point processes, including memorylessness, composability, and transformation of the time axis, that enable exact simulation
  • Implement algorithms to sample event times efficiently using the nhppp R package
  • Conduct conditional simulations, such as simulating events based on diagnoses at specific ages or within defined time intervals
  • Structure the code of a basic discrete-event simulation model

Course Format

This course combines conceptual instruction with guided coding exercises.

Participants will be introduced to discrete-event simulation as a composition of point processes and the properties of nonhomogeneous Poisson point processes that enable efficient simulation. The session will include demonstrations of sampling algorithms and extended hands-on exercises in R, where participants will implement a simple cancer natural history model and practice multiple approaches to event simulation.

Participant Requirements

Participants should have some familiarity with discrete-event simulation and programming in R.

Participants should bring a laptop with R (version 4.0 or later) installed, along with the nhppp and data.table R packages. Familiarity with basic calculus is helpful but not required.