Multistate Models: Analysis from event histories and panel data
Start: 2 March 2020
End: 6 March 2020
Location: Max Planck Institute for Demographic Research (MPIDR), Rostock, Germany
- Hein Putter, Leiden University Medical Center
- Ardo van den Hout, University College London
The life course of individuals can be conceived as a sequence of transitions between different states, for example
- from being healthy to being ill, possibly recovering, and finally to death or
- from living in parental home to living alone, cohabiting with a partner, with or without children, to perhaps living in an institution until death.
The aim of life course analysis is to understand the timing and sequence of transitions as well as the risk factors that accelerate or slow down transitions. Multistate models are the statistical framework to analyze life course patterns and to study and predict resulting population dynamics.
In this 4.5-days course the participants will be introduced to the concepts of multistate models and will learn how to estimate the essential quantities in the two most frequently encountered data situations: Event-histories, for which the exact times of transitions are known, and panel data, where observations are only made in (more or less) regular intervals, leading to interval-censored data.
The course will start with a brief recap of standard survival analysis on which many of the concepts in multistate modeling are based. Moving beyond two-state models the core concepts will be introduced. Besides the estimation of the key parameters, the transition intensities, derived quantities, such as expected lengths of stay in particular states, will be discussed. Selecting and validating well-fitting models, assessing uncertainty of estimates and illustrative presentation of results will also be covered.
There will be an opportunity for participants to present own research ideas within the scope of multistate models.
The course will be a mix of lectures and computer practicals, about 50:50, with about five hours of teaching per day. We will use the statistical software R.
Target audience and prerequisites
The course addresses demographers and researchers from related disciplines such as epidemiology or other social sciences. Participants should be enrolled in a PhD program or have received their PhD.
Participants should have a good working knowledge of standard survival analysis and be familiar with the software R. Students are expected to bring their own laptops with the most recent version of R and an appropriate editor (e.g. Rstudio) installed.
Students’ performance will be evaluated on the basis of a take-home assignment which will be handed out towards the end of the course.
Putter, Fiocco, Geskus (2007). Tutorial in biostatistics: competing risks and multi‐state models. Statistics in Medicine. https://doi.org/10.1002/sim.2712
van den Hout (2016). Multi-State Survival Models for Interval-Censored Data.
Chapman & Hall/CRC Monographs on Statistics and Applied Probability
A list with additional references will be distributed to the participants. Slides and R-code used in the lectures will be made accessible, too.
There is no tuition fee for this course. Students are expected to pay their own transportation and living costs. If you are accepted, MPIDR can provide advice on convenient places to stay in Rostock.
Recruitment of students
- Applicants should either be enrolled in a PhD program or have received their PhD.
- A maximum of 20 students will be admitted.
- The selection will be made by the MPIDR based on the applicants’ scientific qualifications
How to apply
- Applications should be sent by email to the MPIDR (address below). Please begin your email message with a statement saying that you apply for course IDEM 104 – Multistate Models. You also need to attach the following items integrated in *a single pdf file*: (1) A two-page curriculum vitae, including a list of your scholarly publications. (2) A one-page letter from your supervisor at your home institution supporting your application. (3) A two-page statement of your research and how it relates to the course. Please include a short description of your knowledge of survival analysis and your fluency in R. At the very end of your research statement, in a separate paragraph, please confirm that, if admitted, you will be able to come without financial aid from our side.
- Send your email to Heiner Maier (email@example.com).
- Application deadline is 20 January 2020.
- Applicants will be informed of their acceptance by 31 January 2020.
- Applications submitted after the deadline will be considered only if space is available.