← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 6583 · 生成模型:GPT-6

个体单调性约束潜在结果

理解等级:高级

二元处理与结果下,用于这些因果概率公式的无伤害单调性是对几乎每个个体u均有Y_1(u)≥Y_0(u)。它排除处理反而把1变成0的个体。观测条件均值或平均处理反应递增都不足以排除这类个体,不能替代该假设。

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

      Monotonicity is an example of a powerful level 3 assumption. Monotonicity is the simple assumption that the relationship between a cause X and an outcome Y is monotonic: E (Y |X =x ) either never increases or never decreases as x increases. Note that linearity is a special case of monotonicity.

    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:摘录中关于个体单调性约束潜在结果的定义、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:Monotonicity is an example of a powerful level 3 assumption. Monotonicity is the simple assumption that the relationship between a cause X and an outcome Y is monotonic: E (Y |X =x ) either never increases or never decreases as x increases. Note that linearity is a special case of monotonicity.