Robust Decomposition & Sensitivity Analysis

← Back to Domain-Specific Families

Abstractions about component separation, robust sample consensus, variance-based sensitivity, and statistical methods for structured or dependent observations.

5 abstractions in this family — domain-specific abstractions that sit near one another in structural-signature space (k-means over structural-signature embeddings). Each is shown with its short description.

• Bartlett's method — A power-spectrum estimator that averages periodograms from non-overlapping equal-length segments to reduce variance at the cost of frequency resolution.
• Dependent component analysis — A blind-source-separation method recovering mutually independent groups of components while allowing dependence among members of each group.
• Random sample consensus — Estimate a model under substantial outlier contamination by repeatedly fitting random minimal subsets, scoring each hypothesis by thresholded consensus support, and refining the best supported model.
• Rodger's method — A post-hoc multiple-comparison framework selecting orthogonal contrasts while controlling the expected proportion of false rejection decisions.
• Variance-based sensitivity analysis — A global sensitivity method decomposing model-output variance into first-order and interaction contributions from uncertain inputs, commonly summarized by Sobol' indices.