News & Events

Advanced course on Pharmacometrics modeling with Monolix @PAGE2022

on 11 April 2022

Advanced course on Pharmacometrics modeling with Monolix

Tuesday 28 June 2022 – entire day

This training will give you the keys to move your modeling skills to the next level. It is designed for modelers with already a good knowledge of Monolix, who wish to i) better understand the methods, the algorithms and their options, ii) learn how to implement complex models, iii) improve their strategy for model building.
The course will consist of both lectures and interactive hands-on sessions with Monolix. A dedicated Q&A session is also included.

Temporary licences for the MonolixSuite will be provided. This is an in-person workshop taking place in Ljubljana (Slovenia), as a satellite event of the PAGE conference.

Agenda

I New methods and good practices for model building

  • Model diagnostics
    • interpretation of diagnostic plots
    • performing unbiased statistical tests
    • defining and using the shrinkage correctly
  • Strategies for efficient model building
    • automatic covariate search using COSSAC
    • automatic model building using SAMBA

II Model implementation

  • Complex PK models
  • Joint modelling of PKPD data
    • sequential and simultaneous approaches
    • modelling different types of PKPD data (continuous, categorical, count, time to event)

III Methods and algorithms

  • SAEM: theory and practice
    • settings: number of iterations, automatic stopping rule, number of chains
    • simulated annealing
    • convergence assessment
    • handling parameters without variability
  • MCMC: transition kernels, automatic stopping rule
  • Estimation of the Fisher Information Matrix: linearization versus stochastic approximation
  • Estimation of the likelihood function: linearization versus importance sampling Monte Carlo.
  • Bayesian estimation of the population parameters

Main instructor

Prof. Marc Lavielle is a Research Director at Inria.
Marc Lavielle is a statistician specialized in computational statistics and healthcare applications. Marc is responsible for supervising the collaborative research program between Inria and Lixoft. He created and directed the Monolix team at Inria. Marc Lavielle is also the creator of the SAEM estimation algorithm, which was a breakthrough for the study of mixed effect models in drug development.
Marc Lavielle is also a member of the French High Council of Biotechnologies, where he is promoting and studying the application of sound statistical methods to evaluate health and environmental risks related to Genetically Modified Organisms.

 

Price per person

Industry: 600€ – Academia: 300€
The prices are VAT exempted, it will be added if applicable.

Location

Cankarjev Dom cultural and congress centre, Ljubljana, Slovenia (same venue as the conference).

Workshops @PAGE registration form


Jonathan CHAUVINAdvanced course on Pharmacometrics modeling with Monolix @PAGE2022