.png){kind=icon}
.png){kind=icon}
Return
Inferential statistics and direct versus inverse problems
.png){kind=icon}
PRE
AI
DOI：10.1093/aje/kwaf064.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Statistical methods are fundamental to science. However, scientists routinely misinterpret P-values, confidence intervals, and other statistical metrics. This partly results from a lack of clar-ity around core concepts in statistical reasoning. These includeideas about the structure of scientific arguments, as well as the assumptions involved in constructing statistical measures.1For many fields, there is an important distinction betweenproblems that can be classified as direct or inverse. Theseproblems relate to the foundations of statistical inference. Here,we explain the structure of direct and inverse problems, connectthem to inductive and deductive reasoning, and comment on how understanding these issues can bring clarity to the interpretationof statistical results
Keywords:
statistical inference
propositional logic
inverse problems
direct problems
epidemiologic methods
Journal
IF:
4.8
Papers:
9.9K
Citations:
3.7W