[Feature Request]: temporal networks inlcuding multilevel vector autoregressive modelling

[tomtomme]

Description

network analysis: temporal networks inlcuding multilevel vector autoregressive

Purpose

To model directed associations and self-loops in panel and time series data

Use-case

Wide use in temporal associations in longitudinal data with a particular use-case in ecological momentary assessments

Is your feature request related to a problem?

no

Is your feature request related to a JASP module?

Network

Describe the solution you would like

Statistical and graphical features as implemented in the R packages bootnet, qgraph, graphicalVAR, mlVAR, gimme and mgm

Describe alternatives that you have considered

Doing this in RStudio is possiple but not comfortable.

Additional context

An application is described in detail in https://doi.org/10.1177/21677026211017839

This feature request is from a colleague :)

[TarandeepKang]

Morning Team,

I would like to second this request. The above linked paper provides a nice application of the method, but the actual computations and R packages are discussed here:

Bringmann, L. F., Vissers, N., Wichers, M., Geschwind, N., Kuppens, P., Peeters, F., Borsboom, D., & Tuerlinckx, F. (2013). A network approach to psychopathology: New insights into clinical longitudinal data. PloS One, 8(4), e60188. https://doi.org/10.1371/journal.pone.0060188

Epskamp, S., Waldorp, L. J., Mõttus, R., & Borsboom, D. (2018). The Gaussian Graphical Model in Cross-Sectional and Time-Series Data. Multivariate Behavioral Research, 53(4), 453–480. https://doi.org/10.1080/00273171.2018.1454823

Epskamp, S. (2020). Psychometric network models from time-series and panel data. Psychometrika, 85(1). https://doi.org/10.1007/s11336-020-09697-3

Best,

Tarandeep

[TarandeepKang]

Hi Team,

I would just like to reiterate that this would indeed be a very useful feature, and to bring new piece of literature to the discussion, on a newly developed pair of methods to test group differences in Multilevel Vector Autoregressive Models:

Haslbeck, J., Epskamp, S., & Waldorp, L. (2023). Testing for Group Differences in Multilevel Vector Autoregressive Models. PsyArXiv. https://doi.org/10.31234/osf.io/dhp8s

as implemented in the mnet package:

https://cran.r-project.org/web/packages/mnet/index.html

[TarandeepKang]

Hi team,

I would just like to draw your attention to a new paper in longitudinal network analysis:

Hoekstra, R. H. A., Epskamp, S., Nierenberg, A. A., Borsboom, D., & McNally, R. J. (2024). Testing similarity in longitudinal networks: The Individual Network Invariance Test. Psychological Methods. Advance online publication. https://doi.org/10.1037/met0000638

It develops a method to test for "(in)equality between idiographic network structures. INIT extends standard model comparison techniques from Structural Equation Modeling (SEM) to network psychometrics and can be used to test if the same network structure holds across different individuals or if the same network structure holds for one individual over time. By inspecting model selection criteria such as the AIC and BIC, INIT provides an easy way to compare idiographic network structures."

The package is available here: https://github.com/RiaHoekstra/INIT