知识卡片
单调性把响应者比例变成差
令p1=P(Y_1=1)、p0=P(Y_0=1)。在个体无伤害单调性下,PNS=P(Y_1=1,Y_0=0)=p1−p0,因为被排除的“反向响应”类型概率为零。没有该假设,均值差只是正向与反向响应比例之差。
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
- · 识别与因果层级
原书出处 · 1 处
- · 查看原书依据
PNS = P (YX=1 = 1) – P (YX=0 = 1) We can estimate these level 2 estimands from level 2 data, such as a randomized experiment. And, of course, if we only have observational data, we can use backdoor or front-door adjustment or another identification strategy to infer P (YX=0 =1) and P (YX=1 =1) from that data. We could derive these estimands by
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:摘录中关于单调性把响应者比例变成差的定义、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:PNS = P (YX=1 = 1) – P (YX=0 = 1) We can estimate these level 2 estimands from level 2 data, such as a randomized experiment. And, of course, if we only have observational data, we can use backdoor or front-door adjustment or another identification strategy to infer P (YX=0 =1) and P (YX=1 =1) from that data. We could derive these estimands by