RelationalStats

Python toolkit for applied and statistical social network analysis.

RelationalStats provides practical APIs for workflows inspired by R sna, ergm, tergm / stergm, and linkprediction, while keeping methodological boundaries explicit: equivalence, approximation, validation status, and future work are documented instead of hidden.

Package status

• Published package: relationalstats
• Current public release: 0.1.0a1
• Runtime: Python >=3.10
• License: MIT
• Main areas: link prediction, QAPLogit, ERGM approximation, STERGM approximation

Documentation sections

Package READMERead the public package overview, installation flow, quick usage examples, tests, and release notes.Documentation IndexBrowse the synchronized documentation snapshot from the RelationalStats repository.Link PredictionExplore proximity metrics, ProxFun, metric formulas, validation notes, and scalability boundaries.QAPReview QAPLogit, binary dyadic outcomes, node-label permutations, formulas, and R validation plans.ERGM ApproximationUnderstand the initial dyadic-logistic ERGM-inspired approximation, terms, GOF, and limitations.STERGM ApproximationReview separable temporal dyadic construction, formation and dissolution approximations, and validation notes.MethodologyTrack equivalence versus approximation, reproducibility, validation boundaries, and transparent claims.ExamplesRun public, synthetic, reusable examples for link prediction, QAP, ERGM, and STERGM workflows.NotebooksReview the public notebook policy and keep private, solved, or non-public material out of the repository.ReleasesTrack documentation releases for the GitHub Pages version of RelationalStats.

Quick install

pip install relationalstats

For local development from the source repository:

python3.10 -m venv .venv
source .venv/bin/activate
python -m pip install --upgrade pip
python -m pip install -e ".[dev,test]"

Repository

• Source repository: HubertRonald/relationalstats
• PyPI package: relationalstats