METHOD · mth_its · v1
Interrupted Time Series Regression
中断时间序列
Estimates level/trend changes after an intervention time using repeated outcome observations; causal interpretation depends on the counterfactual trend and competing events.
方法身份
- 规范名称
- Interrupted Time Series Regression
- 别名
- ITS、interrupted time series、segmented regression
- 方法族
- quasi-experimental time-series
以上名称和分类来自此版本资料卡,不代表方法之间存在已核验的关系或适用性。
适用问题
- population-level intervention at known time
数据条件
- regular time series
- known intervention timing
- enough pre/post observations
识别与模型假设
- no concurrent event fully explains change
- pre-trend is useful counterfactual
- autocorrelation/seasonality handled
常见失败
- too few pre-periods
- ignoring autocorrelation/seasonality
- attributing all post-change to intervention
不能支持的结论
- causal effect under uncontrolled concurrent shocks
- individual treatment effects from aggregate series
诊断与核对
- autocorrelation
- seasonality
- placebo interruption dates
- alternative functional forms/control series
可比较的替代方法
- controlled ITS
- DiD
- Synthetic Control
软件与实现
- R
- Stata
来源与参考
- Interrupted time series regression for the evaluation of public health interventions: a tutorial
James Lopez Bernal; Steven Cummins; Antonio Gasparrini · 2017
打开原始来源 ↗
来源链接用于核对。来源可访问不代表已获全文或代码再分发授权。
对象关系
关系连接明确的对象版本。前置知识指学习或实现知识;搭配关系是编辑建议,尚未统计真实使用频率。
此对象版本尚未收录明确版本的关系。
关联研究主题
这些旧版关系没有绑定方法版本,只保留供回溯。当前版本的明确版本关系见“对象关系”;编辑建议不代表已验证适用性。
Public Health Intervention Evaluation →
Population intervention at known time
编辑关联 · 待领域复核Climate Adaptation and Urban Resilience →
Time-series intervention evaluation
编辑关联 · 待领域复核