Group Iterative Multiple Model Estimation (GIMME)
The GIMME algorithm is available from the continually maintained R package, gimme. Please select specific topics from the “Tutorial” menu.
The program can be installed directly from R. In addition to the official CRAN documentation and the tutorial topics covered here, please see the “Papers” page of this website for 2 published tutorials.
Program developers are invited to submit changes to our GitHub repository.
The Basics
- GIMME can search for patterns of relations according to three models:
- Unified SEM: both lagged and contemporaneous relations are directed
- Vector Autoregression: lagged relations are directed, contemporaneous relations are bidirectional correlations among residuals
- Hybrid-VAR: lagged relations are directed; contemporaneous relations can be either directed among the variables or bidirectional among residuals
- Missing data is not a problem. Simply put “NA” where the data are missing. However – if there are more rows of missing data than there are rows with data, then convergence problems may arise.
- Heterogeneous data is not a problem:
- No “group” or “common” structure will be forced unless it truly describes the majority.
- Individual-level nuances will surface after a group or common structure is fit (provided one exists).
- If desired, subgroups of individuals with similar patterns of effects will be generated to aid the researcher in finding similar patterns among the varied individual models.
- Works well with as little as 3 or as many as 25 variables (between 5 and 15 is recommended).
- Can be freely downloaded by installing the package “gimme” in R.