Skip to content
Carmelics
Topics
Thinkers
Changes
Contributors
Loading account…
Statements
321,452
Perspectives
108,905
Topics
42
Home
/
Original
/
inverse
See Original
Inverse View
It is not the case that Fiducial probability can give classical statistics an epistemic status
?
Set your confidence on the premises below to see your aggregate.
Reasons For
2 perspectives
Reason for 1 of 2
?
1.
Fiducial distributions lack coherence: they can assign probabilities that violate the axioms of probability when applied to different parameterizations of the same model.
?
How convincing is this?
Think about whether this reason is strong or weak
2.
A probability assignment that fails coherence cannot legitimately ground epistemic credence in hypotheses, as Lindley demonstrated via fiducial contradictions in the 1950s.
?
How convincing is this?
Think about whether this reason is strong or weak
Reason for 2 of 2
?
1.
Fisher's fiducial argument illicitly converts a frequency statement about estimators into a posterior-style statement about parameters without epistemic justification.
?
How convincing is this?
Think about whether this reason is strong or weak
2.
The logical gap between 'the procedure traps the true value 95% of the time' and '95% probability the parameter lies here' requires exactly the prior-free inference fiducial claims to achieve, making the derivation circular.
?
How convincing is this?
Think about whether this reason is strong or weak
Reasons Against
1 perspective
Reason against
?
Fiducial probability derives a probability assignment over hypotheses without assuming a prior probability over statistical hypotheses at the outset
?
How convincing is this?
Think about whether this reason is strong or weak
Next step
Based on where you are in your exploration
Strongest counterpoint
Explore the most compelling reason on the other side.