...
首页> 外文期刊>Economics & philosophy >IMPLEMENTATION-NEUTRAL CAUSATION
【24h】

IMPLEMENTATION-NEUTRAL CAUSATION

机译:实施中性因果关系

获取原文
获取原文并翻译 | 示例
           

摘要

The most basic question one can ask of a model is `What is the effect on variable y(2) of variable y(1)?' Causation is `implementation neutral' when all interventions on external variables that lead to a given change in y(1) have the same effect on y(2), so that the effect of y(1) on y(2) is defined unambiguously. Familiar ideas of causal analysis do not apply when causation is implementation neutral. For example, a cause variable cannot be linked to an effect variable by both a direct path and a distinct indirect path. Discussion of empirical aspects of implementation neutrality leads to further unexpected results, such as that if one variable causes another the coefficient representing that causal link is always identified.
机译:一个模型可以问的最基本的问题是“对变量y(1)的变量y(2)有什么影响?”当所有导致y(1)发生给定变化的外部变量干预均对y(2)产生相同影响时,因果是“实现中立”,因此明确定义y(1)对y(2)的影响。当因果关系是实施中立的时候,因果关系分析的熟悉概念将不适用。例如,原因变量不能通过直接路径和独特的间接路径链接到效果变量。对实现中立性的经验方面的讨论会导致进一步出乎意料的结果,例如,如果一个变量导致另一个变量,则代表因果关系的系数将始终被识别。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号