知识卡片
不确定性不只来自采样
即使目标量已识别,有限样本估计仍有误差;还可能存在机制、参数化或选择假设的不确定性。显著性检验只回答特定统计问题,应结合效应量、区间和决策成本,不把p值通过当成行动有实际价值。
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
Thus, traditional non-causal ways of quantifying statistical association become measures of causal association once we use them in an interventional context. And when we fit these regression models to data, we can still use all the traditional regression methods for significance testing (Wald tests, F-tests, likelihood ratio tests, etc.).
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:摘录中关于不确定性不只来自采样的定义、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:Thus, traditional non-causal ways of quantifying statistical association become measures of causal association once we use them in an interventional context. And when we fit these regression models to data, we can still use all the traditional regression methods for significance testing (Wald tests, F-tests, likelihood ratio tests, etc.).