Copulas for covariate simulation in R – Leiden University

PAGE 2024 satellite workshop

Copulas can be used to estimate covariate distributions and relationships and simulate new patient covariate sets. Recently, we proposed this method for covariate simulation in a previous PAGE talk and recent paper. Sign up to find out what the copula is and to learn how to use it on your own data.

By the end of this workshop you will

• have a theoretical understanding of (vine) copulas;
• understand the estimation process of copulas on data;
• be able to fit a copula on data and simulate from the copula;
• be able to visualize and evaluate the fit of the copula and simulated populations;
• be able to use simulated covariates in a population PK/PD simulation

Prerequisites: We expect you to have previous experience in R and bring your laptop with an installation of R.
Date: Tuesday, 25th June, 2024
Time: 13:00 – 17:00
Fees: €150 for industry, €75 for academics
Registration: link

More information
Simulation is a core element of pharmacometric research. While PK and (PK-)PD profiles can be simulated from models, covariate simulation requires the use of underlying patient data. Copulas allow for characterizing dependence structures between patients’ covariate sets and subsequent simulation. These copulas can be shared among researchers, facilitating the creation and sharing of patient covariate simulators. This workshop is intended as a step-by-step guide for understanding the concept of and applying copulas in pharmacometric research.

We expect you to have previous experience in R and bring your laptop with an installation of R. We will be using the package ‘rvinecopulib’ and will provide a toy dataset. It is recommended to bring your own data, so by the end of the workshop you will have a copula model of your data, and the accompanying R scripts for estimation and simulation of the copula.

Organizers
The workshop is organized and taught by dr. Laura B. Zwep, Yuchen Guo, MSc, dr. Tingjie Guo and dr.  J.G. Coen van Hasselt, all associated to the Leiden Academic Centre for Drug Research, Leiden University, the Netherlands. At this academic institute, research and teaching go hand-in-hand, as such, we aim to provide an engaging workshop of high quality.

.