Training and Events
Using the Relational Event Model (REM) to investigate patterns in ordered/timed sequences of actions
|The University of Edinburgh|
Dr Mark Tranmer, Professor of Quantitative Social Sciences, The University of Glasgow
The University of Edinburgh
View in Google Maps (EH3 9DR)
The course will explain how the Relational Event Model (REM) may be used to investigate patterns in ordered or timed sequences of actions.
We begin by giving some examples for which ordered sequences of timed data may be collected, including patterns of behaviour of individuals over time, and interactions in a network of individuals over time.
We then introduce the REM in the context of other related methods, such as survival analysis, and also in the context of other ways of looking at sequences of actions, such as sequence analysis.
We explain the importance of taking into account the way in which the ordered or timed data were collected, and the actions that were observable at each point in the sequence, when analysing it, and explain how this can be achieved.
Next, we explain how an R package called informR can enable us to prepare ordered or timed data for analysis with a REM. Data preparation with InformR allows us to take into account the way in which the data were collected, and the actions that are observable at each point in the sequence. InformR also allows us to create “sequence statistics” to allow us to investigate particular patterns in the timed or ordered sequence of actions that may be of particular substantive interest.
Finally we explain how relevent, an R package, can be used to fit REMs to ordered or timed sequence data. We give examples, and explain how the results of a REM analysis that has been carried out using relevent can be interpreted.
One Day Course with registration from 9.30am
Course will run from 10am-4.30pm
Intermediate (some prior knowledge)
£30 per day for UK/EU registered students
Website and registration
Related publications and presentations