4 Software Projects
4.1 semnova
semnova is an R package for analyzing data from repeated measures experimental designs. semnova implements latent repeated measures analysis of variance (RM-ANOVA) which is a structural equation modeling based alternative to tradtional RM-ANOVA. Traditional RM-ANOVA is a widely used statistical tool in the field of psychology and social sciences. RM-ANOVA can analyze data such as test scores, questionnaire items, reaction times, attitudes, characteristics, or motives. Oftentimes, the construct of interest cannot be observed directly and the aforementioned measures contain measurement errors. Latent RM-ANOVA can include multiple indicators that measure the same latent construct of and prune of measurement error. Latent RM-ANOVA can furthermore be used to examine interindividual differences in main and interaction effects of experimental factors and introduces a whole lot of other advantages from the structural equation modeling framework.
Check out:
4.2 subgroupsem
subgroupsem is an R package for mining exceptional models. Althoug the package was originally developed for mining structural equation models, it can be used with any kind of statitical model and external packages as it optimizes a user-defined function. Given a dataset and a set of covariates or group variables, subgroupsem uses algorithms from the data mining framework to find exceptional sub-samples with respect to the user-defined criterion. subgroupsem itself is only a wrapper for the python module pysubgroup authored by Dr. Florian Lemmerich and makes use of the reticulate package to interface R with pysubgroup. Together with Dr. Lemmerich, we demonstrated how to use model mining alogorithms to structural eqaution models. Check out the corresponding article below.
Check out: