- Organisation & People
- Research & Publications
- PhD Programme
- PhD Roadmap, TSP & Diploma
- PhD Graduations
- News & Events
This is a four-day course on how to study dynamics in intensive longitudinal data, such as ambulatory assessments (AA), experience sampling method (ESM) data, ecological momentary assessments (EMA), real time data capture, observational data or electronic daily diaries. We provide a tour of diverse modeling approaches for such data and the philosophies behind them, as well as practical experience with these modeling techniques using different software packages (including R and Mplus).
Technological developments such as smartphones, activity trackers, and other wearables have made it relatively easy to obtain many repeated measurements per person in a relatively short period of time. In response to these measurement innovations, there is a surge of statistical modeling innovations that are designed to handle the unique challenges of such intensive longitudinal data and uncover the most valuable and meaningful properties of these data. A particular appealing aspect of such data is that the observations are ordered in time, thereby allowing us to study the dynamic relationships between variables over time. Moreover, when data come from multiple cases (e.g., individuals or dyads), we can also study the similarities and differences in the means, variability and dynamics of these cases.
On day 1 of this four day course we begin with grounding ourselves in N=1 time series analysis as it has been employed for decades in econometrics. We will cover basic topics such as the ARIMA model, the autocorrelation function, stationarity, unit roots and trends. Building on this basis, we will discuss N>1 extensions and dynamics multilevel modeling during day 2, and emphasize the importance of separating within-person dynamics from between-person differences. On day 3 we will discuss measurement issues, including some modeling solutions such as models that account for measurement error. Additionally, we discuss dynamic network analysis. On day 4, we discuss continuous time modeling and changes in dynamics.
For this course some knowledge of multilevel analysis and Bayesian statistics is preferable, but not required. Also, experience with R and/or Mplus is useful, but is not mandatory.
Please note that there is always the possibility that we have to change the course pending COVID19-related developments. The exact details, including a day-to-day program, will be communicated 6 weeks prior to the start of the course.