| Session | |
WS-T43: Advanced Modeling of Relational Events in R Using goldfish.latent
| |
| Session Abstract | |
|
This workshop provides an advanced introduction to `goldfish.latent`, an R package that extends relational event modeling by incorporating latent variable models. Participants will learn to model actor heterogeneity through the package's implementation of random effects powered by Stan. Practical examples and hands-on exercises will guide attendees through model specification, estimation, and interpretation, enabling them to apply these advanced methods to their relational event data. A particular focus will be given to analyzing multiple sequences as a case study for using random effects, highlighting the package's flexibility in handling complex relational event structures. Prerequisites: Participants should be familiar with R and the `goldfish` package. Those new to goldfish are encouraged to attend the introductory “Modeling Relational Events in R Using goldfish” workshop. What to Bring: • A laptop with the following installed: o R statistical computing system o Stan (via `cmdstanr` or `rstan`) o `goldfish` and `goldfish.latent` packages with all dependencies • Installation links: o goldfish.latent: https://github.com/snlab-ch/goldfish.latent o Stan: https://mc-stan.org/cmdstanr/ References: • Stadtfeld, Christoph, and Per Block. 2017. “Interactions, Actors, and Time: Dynamic Network Actor Models for Relational Events.” Sociological Science 4 (14): 318–52. https://doi.org/10.15195/v4.a14. • Uzaheta, Alvaro, Viviana Amati, and Christoph Stadtfeld. 2023. "Random Effects in Dynamic Network Actor Models." Network Science 11(2): 249-266. https://doi.org/10.1017/nws.2022.37. Length: 3 hours Participants: 30 | |
| Presentations | |
Advanced Modeling of Relational Events in R Using goldfish.latent This workshop provides an advanced introduction to `goldfish.latent`, an R package that extends relational event modeling by incorporating latent variable models. Participants will learn to model actor heterogeneity through the package's implementation of random effects powered by Stan. Practical examples and hands-on exercises will guide attendees through model specification, estimation, and interpretation, enabling them to apply these advanced methods to their relational event data. A particular focus will be given to analyzing multiple sequences as a case study for using random effects, highlighting the package's flexibility in handling complex relational event structures. Prerequisites: Participants should be familiar with R and the `goldfish` package. Those new to goldfish are encouraged to attend the introductory “Modeling Relational Events in R Using goldfish” workshop. What to Bring: • A laptop with the following installed: o R statistical computing system o Stan (via `cmdstanr` or `rstan`) o `goldfish` and `goldfish.latent` packages with all dependencies • Installation links: o goldfish.latent: https://github.com/snlab-ch/goldfish.latent o Stan: https://mc-stan.org/cmdstanr/ References: • Stadtfeld, Christoph, and Per Block. 2017. “Interactions, Actors, and Time: Dynamic Network Actor Models for Relational Events.” Sociological Science 4 (14): 318–52. https://doi.org/10.15195/v4.a14. • Uzaheta, Alvaro, Viviana Amati, and Christoph Stadtfeld. 2023. "Random Effects in Dynamic Network Actor Models." Network Science 11(2): 249-266. https://doi.org/10.1017/nws.2022.37. Length: 3 hours Participants: 30 | |