Decision-Oriented Two-Parameter Fisher Information Sensitivity Using Symplectic Decomposition

The eigenvalues and eigenvectors of the Fisher Information Matrix (FIM) can reveal most least sensitive directions a system it has wide application across science engineering. We present symplectic variant eigenvalue decomposition for FIM extract sensitivity information with respect to two-parameter conjugate pairs. approach decomposes onto an even-dimensional basis. This structure additional between pairs, otherwise concealed in orthogonal basis from standard decomposition. proposed be applied naturally paired distribution parameters, or decision-oriented pairing via regrouping re-parameterization FIM. It used tandem offer insights into analysis at negligible extra cost. Supplementary materials this article are available online.