知识卡片
工具变量不自动识别总体效应
工具的图条件一般不足以非参数识别总体ATE。线性恒定效应模型等额外条件可支持特定估计;二元工具和处理在独立、排除、相关及个体单调性下通常针对服从者局部平均效应。必须说明最终目标与额外假设。
这张卡片看不懂?
这张卡片已通过人工审查,不再接受失败反馈。
已有反馈 · 0 条
还没有人提交反馈。
在整书笔记中的位置
原书出处 · 1 处
- · 查看原书依据
additional parametric assumptions are needed. DoWhy, by default, makes a linearity assumption. With a linear assumption, you can derive the ATE as a simple function of the coefficients of linear models of outcome and the treatment given the instrument. DoWhy does this by fitting linear regression models.
参考来源
- 位置:《Causal AI》第11章《构建因果推断工作流》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Causal AI/_epub-src/OEBPS/Text/chapter-11.xhtml)
- 结论依据:摘录中关于工具变量不自动识别总体效应的定义、比较或条件支撑卡片;原书表述不严谨的范围已在覆盖记录逐项限定,不把案例结果外推为普遍结论。
- 原始内容:additional parametric assumptions are needed. DoWhy, by default, makes a linearity assumption. With a linear assumption, you can derive the ATE as a simple function of the coefficients of linear models of outcome and the treatment given the instrument. DoWhy does this by fitting linear regression models.