知识卡片
多类别效应需要明确基线
当处理有a、b、c等取值,可报告b对a、c对a或其他明确的成对对比。把多个类别并成“非a”还需要指定其中各处理版本及混合权重,否则干预对象不完整。不同基线和混合策略可以对应不同目标量。
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
- · 干预与因果效应
原书出处 · 1 处
- · 查看原书依据
When the cause is categorical, the ATE requires choosing which levels of the cause you want to compare. For example, if X has possible outcomes {a, b, c}, you might select “a” to be a baseline, and work with two ATEs, E (YX = b – YX = a) and E (YX = c – YX = a). Alternatively, you may choose to work with
参考来源
- 位置:《Causal AI》第07章《干预与因果效应》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Causal AI/_epub-src/OEBPS/Text/chapter-7.xhtml)
- 结论依据:摘录中关于多类别效应需要明确基线的定义、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:When the cause is categorical, the ATE requires choosing which levels of the cause you want to compare. For example, if X has possible outcomes {a, b, c}, you might select “a” to be a baseline, and work with two ATEs, E (YX = b – YX = a) and E (YX = c – YX = a). Alternatively, you may choose to work with