知识卡片
预测比较须评价增量信息
比较只用Z的模型与用X和Z的模型,可探索X是否提供额外预测信息。需要合适损失、充分模型能力及独立评价数据;两模型分数接近可能来自低功效或欠拟合,不能一般证明X⊥Y|Z。
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
原书出处 · 1 处
- · 查看原书依据
M1 predicts Y using Z as a predictor. M2 predicts Y using X and Z as a predictor. Predictors can have dimensions greater than one. If X ⊥ Y | Z, then any X has no predictive information about Y beyond what is already provided by Z. So you can test X ⊥ Y | Z by comparing the model predictive accuracy of M2 to
参考来源
- 位置:《Causal AI》第04章《用因果约束检验DAG》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Causal AI/_epub-src/OEBPS/Text/chapter-4.xhtml)
- 结论依据:摘录中关于预测比较须评价增量信息的定义、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:M1 predicts Y using Z as a predictor. M2 predicts Y using X and Z as a predictor. Predictors can have dimensions greater than one. If X ⊥ Y | Z, then any X has no predictive information about Y beyond what is already provided by Z. So you can test X ⊥ Y | Z by comparing the model predictive accuracy of M2 to