Insight & Inference for DVARS: 3-part variance decomposition, gives units & p-values to DVARS measures as well as spatiotemporal rs-fMRI diagnostic.
-
Updated
Mar 23, 2024 - MATLAB
#
variance-decomposition
Here are 6 public repositories matching this topic...
A diagnostic control paradigm for activation measurements in transformer language models. Cross-replay separates text-bound from architecture-bound components by replaying generated sequences through intact and perturbed model variants.
language-models pythia interpretability variance-decomposition linear-probing permutation-test detrended-fluctuation-analysis gpt-2 activation-analysis representational-similarity transformer-interpretability gpt-j centered-kernel-alignment mechanistic-interpretability attention-analysis cross-replay procrustes-distance
-
Updated
May 17, 2026 - Jupyter Notebook
Global sensitivity analysis — Song 2016 Shapley effects, Rabitti-Tzougas 2025 CLH subsampling, exposure-weighted variance decomposition
-
Updated
Apr 5, 2026 - Python
End-to-End Python implementation of the Mesoscopic Structural Risk-Navigation System from Shao, Yang & Zhang (2026). Estimates rolling QVAR models, extracts multiscale network backbones via Disparity Filter, enumerates 13 directed triadic motifs (30 node orbits), and optimizes a Minimum Structural Similarity Portfolio.
python topology jupyter-notebook network-science portfolio-optimization time-series-analysis computational-finance graph-isomorphism variance-decomposition systemic-risk financial-networks disparity-filter risk-spillovers triadic-motifs connectedness-analysis quantile-var quantile-econometrics mesoscopic-analysis backbone-extraction node-orbits
-
Updated
May 2, 2026 - Jupyter Notebook
Deterministic engine for 'why did this metric change?' — decomposes the move over a hypothesis tree you define, reconciles every step, and abstains instead of guessing.
python analytics business-intelligence simpsons-paradox variance-decomposition root-cause-analysis llm-tools
-
Updated
Jun 20, 2026 - Python
Measures what language models say about a brand and reports what the sample actually supports. Wilson intervals, a variance split that says whether more questions or more repeats buy your precision, and a hash-chained ledger that seals its own length.
python bootstrap cli statistics geo confidence-intervals sampling measurement reproducibility variance-decomposition ai-search tamper-evident perplexity brand-monitoring append-only-log llm byok anthropic cost-tracking generative-engine-optimization
-
Updated
Aug 4, 2026 - Python
Improve this page
Add a description, image, and links to the variance-decomposition topic page so that developers can more easily learn about it.
Add this topic to your repo
To associate your repository with the variance-decomposition topic, visit your repo's landing page and select "manage topics."