In chjackson/msm: Multi-State Markov and Hidden Markov Models in Continuous Time

options(digits=3)

Course overview {-}

This course gives an introduction to the use of the msm R package for fitting and predicting from multi-state models for intermittently observed data, in typical medical applications.

There is a balance of short talks and practical exercises. In the exercises, you will write your own R code, based on the code demonstrated in the talks, to perform multi-state data analyses and answer applied questions.

All the course material will stay on this website, http://chjackson.github.io/msm/msmcourse,

so you can follow along at your own pace, and finish exercises in your own time.

Rough timings:

• 9.00 - 9.50

  • Part 1. Multistate modelling theory review
  • Part 2. Basic use of msm
  • Exercises

• 9.50 - 10.30

  • Part 3. Prediction from models with msm
  • Exercises

• 10.40 - 11.30

  • Part 4. msm models with covariates
  • Exercises

• 11.30 - 12.30

  • Part 5. msm models with time-dependent covariates
  • Exercises
  • Part 6. Advanced topics (if there is time)

Prerequisites {-}

Computer with R and the msm package installed.

Basic knowledge of R.

Basic statistics / probability, ideally from a medical / biostatistics perspective.

Some knowledge of survival analysis would help.

# automatically create a bib database for R packages
knitr::write_bib(c(
  .packages(), 'bookdown', 'knitr', 'rmarkdown'
), 'packages.bib')

Licence {-}

The material in this course is freely available and redistributable under the terms of the CC-BY licence.

chjackson/msm documentation built on March 3, 2024, 1:05 a.m.