Towards a Theory of Valid Inferential Models with Partial Prior Information
2021
Inferential models (IMs) are used to quantify uncertainty in statistical inference problems, and validity is a crucial property that ensures the IM’s reliability. Previous work has focused on validity in the special case where no prior information is available. Here I allow for prior information in the form of a non-trivial credal set, define a notion of validity and investigate its implications.
Keywords:
- Correction
- Source
- Cite
- Save
- Machine Reading By IdeaReader
20
References
0
Citations
NaN
KQI