FEAR: Flood Exposure and Asset Repricing -- Replication Package
================================================================
Cameron Keith | Econ 66 | Winter 2026

This package contains the data and code needed to reproduce all
regression results in the paper.


REQUIREMENTS
------------
- Stata 17+ with the following packages installed:
    ssc install reghdfe, replace
    ssc install ftools, replace
    ssc install estout, replace
    ssc install bacondecomp, replace
    ssc install drdid, replace
    ssc install csdid, replace
    ftools, compile
    reghdfe, compile


CONTENTS
--------
event_study.do                              Stata do-file (all regressions)
data/clean/regression_panel.csv             Main regression panel (zip x month)
data/clean/election_county_year.csv         County-level presidential vote shares
data/clean/nfip_lomr_deltas.csv             NFIP policy/claim deltas around LOMRs
data/raw/state-disclosure-laws/disclosure_laws.csv   State flood disclosure law indicators
output/results/                             (empty) output directory for results


HOW TO RUN
----------
1. Unzip this archive.
2. Open Stata and cd to the unzipped directory:
       cd "/path/to/replication"
3. Run the do-file:
       do event_study.do
4. Results (CSVs, TeX tables, PNGs) will appear in output/results/.

The do-file uses `global root "."` so it expects to be run from the
replication directory. If you prefer an absolute path, edit line 57.


OUTPUT
------
The do-file produces:
- Event study coefficient CSVs (s05, s06, s06b, s08, s09, etc.)
- Regression tables in TeX and CSV (s02-s16)
- Event study plots (PNG)
- Bacon decomposition data
- Leave-one-out sensitivity results
- Callaway & Sant'Anna estimates
- Log file: output/event_study.log
