FINANCEFIRST FLOOD REPORT | RESEARCH DATA KIT v1.1

Report: https://financefirst.co/reports/flood-risk-insurance-gap
Author: Aisha Shabbir. Human insurance/legal review has not been recorded.
GAO source release: October 1, 2026. FinanceFirst publication: November 11, 2026. Revision/research check: November 11, 2026. Data vintages are different; see methodology.txt.

QUICK START
1. Extract this ZIP into a folder.
2. Open a terminal in that folder.
3. Run: python reproduce.py
Python 3 standard library only. No calculate.py, install step, network access, credentials or external datasets required. Windows can use py reproduce.py if Python is registered as py.
Expected: PASS, combined count 13039240, displayed share 81.3, and six retained costs 106000/28000/10000/16000/18000/58000. See verification/results.json for all checks and payment components.

FOLDER STRUCTURE
reproduce.py
data/raw/gao-table1.csv
data/raw/hypothetical-inputs.csv
data/derived/high-risk-location.csv
data/derived/household-scenarios.csv
data/coverage-audit.csv
assets/ (two explanatory chart SVGs)
sources.csv; claims.csv; methodology.txt; article.html; CHANGELOG.txt

All household inputs are invented. This does not reproduce First Street risk modeling, determine a personal flood probability or adjudicate an insurance claim. The hero illustration is separate from this data kit. Federal/source data are attributed to their publishers. No claim of ownership over proprietary property-level data.
