方法证据记录
Bayesian Panel Event Study
Bayesian Panel Event Study is a causal inference design that estimates dynamic treatment effects around a datable event using panel data, replacing classical frequentist estimation with Bayesian posterior inference. It produces period-by-period effect estimates with full probability distributions, enabling principled uncertainty quantification, regularization of noisy pre-trend coefficients, and probabilistic tests of parallel trends.
源记录
引文逐字复制自方法源记录。这些引文不代表任何层级的验证。
Bayesian Panel Event Study Design
分类方法记录 · regression-model / causal-inference
- Freyaldenhoven, S., Hansen, C., Shapiro, J. M., & Teso, E. (2021). Visualization, Identification, and Estimation in the Linear Panel Event-Study Design. NBER Working Paper No. 29170. National Bureau of Economic Research. · URL
- Jakiela, P. (2021). Simple Diagnostics for Two-Way Fixed Effects. Working Paper. Center for Global Development. · URL
精选声明
声明已持久化到证据分类账中，每个声明都有自己的评估。
尚无精选声明
当分类账中没有声明时，此视图不会自行创建声明评估。
相关方法
从方法图中生成，显示为机器建议的关系 — 不推断任何证据声明。