Data & Analytics
Data Breach Disclosures & Equity Market Reactions
Event-study and difference-in-differences analysis of how U.S. equity markets react around corporate data-breach disclosures.
Role: Analyst / researcher (reproducible R project)
Situation
Cyber incidents are often discussed as operational events; this project measures whether and how markets price public breach disclosures using financial econometrics.
Decision question
Event-study and difference-in-differences analysis of how U.S. equity markets react around corporate data-breach disclosures.
Data and constraints
See method and results below for sample, coverage, and documented limits.
Method
Combined a cited breach-events dataset with daily firm and S&P 500 prices. Ran a classic market-model event study (AR/CAR) and a Callaway- Sant- Anna dynamic ATT pipeline, with documented sample rules, confound handling, and reproducible outputs.
Tools
- R
- tidyquant
- Event study methods
- Callaway- Sant- Anna / did
- OLS
Deliverables
- Event-study scripts and knit-ready report
- CS/DID estimator pipeline
- Figures, tables, and methodology documentation
Outcome / decision value
- Documented run (2026-04-20): mean CAR(0,+10) ? ?0.12% (n = 224 non-confounded); mean AR sharply negative at disclosure day (t = 0).
- CS/DID overall dynamic ATT ? ?0.22 (95% CI [?0.42, ?0.02]); pre-trend p ? 0.91.
- Breach-size association with CAR had the expected negative sign but was not statistically strong in the baseline cross-section (p ? 0.23).