← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 6580.b · 生成模型:GPT-6

后门推导还需一致性与重叠

理解等级:高级

潜在结果写法中,条件可交换性Y_x⊥X|Z、一致性和正值性共同支持调整识别。条件独立不提供未观测处理层的结果数据;若某类Z下从未出现X=x,不能无条件用层内回归填补并称非参数识别。

这张卡片看不懂?

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

      Z is a set of variables called the adjustment set. The summation is shorthand for summation and integration—you sum over discrete variables in the adjustment set and integrate over continuous variables. The adjustment set is defined as fa set of variables that satisfies the backdoor criterion—(1) the set collectively d-separates all backdoor paths from X to Y, and (2) it contains no descendants of X.

    参考来源

    - 位置:《Causal AI》第10章《识别与因果层级》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Causal AI/_epub-src/OEBPS/Text/chapter-10.xhtml) - 结论依据:摘录中关于后门推导还需一致性与重叠的定义、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:Z is a set of variables called the adjustment set. The summation is shorthand for summation and integration—you sum over discrete variables in the adjustment set and integrate over continuous variables. The adjustment set is defined as fa set of variables that satisfies the backdoor criterion—(1) the set collectively d-separates all backdoor paths from X to Y, and (2) it contains no descendants of X.