Panel data contains measurements from multiple subjects measured over multiple time points. Such data can be encountered in many social science applications such as when analysing register data or cohort studies (for example). Often the aim is to perform causal inference based on such observational data (instead of randomized control trials).
This is a companion discussion topic for the original entry at https://ropensci.org/blog/2023/01/31/dynamite-r-package/