知识卡片
预测关系不自动代表因果作用
广告与销量的回归可用于相应分布下的预测,却不能仅凭拟合解释增加广告会造成多少销量。共同因素与行动选择可能影响关系;若目标是干预收益,需要另外的设计或识别论证。
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
- · 机器学习基础
原书出处 · 1 处
- · 查看原书依据
Suppose we are interested in the relationship between sales and advertising. Companies spend millions of dollars on advertising and expect that an increase in advertising expenditure will significantly improve sales. Thus, these two variables might be related. Other examples where two variables might be related are production cost and the volume of production, increase in summer temperature and the home heating cost, or the size
参考来源
- 位置:《Machine Learning Fundamentals: Concepts, Models, and Applications》第02章《机器学习基础》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Machine Learning Fundamentals: Concepts, Models, and Applications/_epub-src/OEBPS/xhtml/13_chapter02.xhtml)
- 结论依据:摘录中关于预测关系不自动代表因果作用的定义、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:Suppose we are interested in the relationship between sales and advertising. Companies spend millions of dollars on advertising and expect that an increase in advertising expenditure will significantly improve sales. Thus, these two variables might be related. Other examples where two variables might be related are production cost and the volume of production, increase in summer temperature and the home heating cost, or the size