Research Notes
Short, visual explanations of the ideas behind my work—designed to be easier to read than a formal methods paper, but still technically meaningful.
What is the Longitudinal Rank-Sum Test?
Why a global rank-based test is useful, how the ranking intuition works, when LRST is appropriate, and how the R package fits into the workflow.
Why higher-order spectra?
Why ordinary spectra can miss nonlinear and non-Gaussian structure, how bispectra and polyspectra reveal richer dependence, and why polyspectral means are useful summaries.
What causal discovery can—and cannot—tell us
What structure-learning algorithms identify, why equivalence classes matter, how latent confounding changes the graph, and why unresolved endpoints matter.
These notes are intentionally visual and conceptual. For code, implementation details, and longer walkthroughs, see Tutorials & Resources.