Sessions

≪Organizer Abstract≫
Structural equation models (SEM) have been used for analysis of complex directional relationships among observed variables. Applications of SEM include analysis of brain connectivity based on fMRI and other types of data (e.g., EEG). However, there are typically hundreds of measured variables, for which conventional SEM methods (based on covariance analysis) are not suitable. Recently, a more straightforward approach to SEM based on component analysis (called generalized structured component analysis) has been proposed to deal with massive data sets by the organizers/speakers of this symposium. In this symposium, we overview the current state-of-the-art of this technique, introduce various applications, and discuss possible desiderata for future expansions.
Advanced Search