Higher-Ed A/G Research Extension A public IPEDS study of institutional closure risk in a validated higher-ed subdomain.

Evidence

This page indexes the reader-facing summaries of the documents behind the locked higher-ed result. The summaries are written for a non-internal audience; the full markdown source documents stay in the project repository for reviewers who need the raw record.

No institution-level scores, no institution-level rankings, and no raw IPEDS data are linked from this page. The aggregated validation tables (cohort-level metrics, fold AUROCs, regime base rates) are summarised in each document’s page.

Start here

Final Lock Note

Plain-language summary of what the higher-ed extension is, what it is not, the validated lane, the headline result, and the Phase-19 hardening evidence.

Read the Final Lock Note →

Claim Boundary Note

Canonical allowed and forbidden claims. Includes the boundary checklist that must accompany any external restatement of the result.

Read the Claim Boundary Note →

Phase 19 Hardening Decision

Integrated decision across the five Phase-19 hardening tracks: cluster bootstrap, temporal multi-fold, ED Financial Responsibility comparator, HCM source-blocked, lead-time analysis.

Read the Phase 19 Decision →

Core validation

Cluster Bootstrap

Institution-level (UNITID) bootstrap of the closure AUROC. Cluster-bootstrap 95 % CI compared with the original analytic Hanley-McNeil interval, plus paired-difference distribution against the strongest tested single-variable baseline.

Read the Cluster Bootstrap Note →

Temporal Folds

Four sliding train/test time-block folds. Joint-weak AUROC stays in [0.77, 0.86] across all four folds; honest record of the Fold-1 baseline tie.

Read the Temporal Folds Note →

Closure Lead-Time

Score distributions at T-1 through T-5 for the closing FASB × Sector 2 institutions, and the top-decile capture rate by years-to-closure.

Read the Lead-Time Analysis →

ED Financial Responsibility Baseline

The federal composite score as a comparator on the descriptive train cohort. Head-to-head test on the locked cohort is not runnable due to a data-coverage gap.

Read the ED FR Baseline Note →

Heightened Cash Monitoring Anchor

Source-blocked in this run; documented as a candidate for future work. No HCM evidence is incorporated into the locked claim.

Read the HCM Anchor Note →

Reproducibility & evidence index

Artifact Index

Inventory of higher-ed phase artifacts marked CURRENT, HISTORICAL, or SUPERSEDED so a reader can tell what is the current evidence and what is preserved as historical record.

Open the Artifact Index →

Evidence Packet Manifest

The 13-item minimum-evidence packet, plus the exclude-from-review-packets list (raw IPEDS files and the full panel CSV are excluded by default).

Open the Evidence Packet Manifest →

Reproducibility Runbook

End-to-end re-run instructions and the lock-evidence pipeline subset (Phases 12.5 → 17) needed to verify the headline result.

Open the Reproducibility Runbook →