9 problems

Half-coherence conjecture. The main reason that conformal prediction handles the vacuous-prior case more efficiently than other imprecise-probability frameworks is that CP enjoys h…

Impossibility conjecture. There is no reasonable estimator such that

An inferential model (IM) produces a coherent lower and upper probability and has strong validity, meaning that its plausibility contour is stochastically no smaller than uniform u…

An inferential model (IM) is constructed from a statistical model with parameter $Theta$ and sampling distribution $\mathsf{P}_{Y|\theta}$; its plausibility contour is driven by th…

Let $\mathcal{Z}$ be the set of precise models under consideration, and define the lower updated expectation by … whenever the set over which the infimum is taken is nonempty. Limi…

A conditional lower prevision is called 2-convex or 2-coherent when it satisfies the corresponding consistency property for combinations of at most two gambles; its natural extensi…

Unconditional convergence conjecture. This limit should exist without imposing any conditions on $\underline{Q}$:

Let ${\cal P}$ be a set of distributions on ${\cal X}\times{\cal Y}$, and let $\delta_{\mathrm{prior}}$ denote the a priori minimax-optimal decision rule. A partition…

Let ${\cal P}$ be an arbitrary set of distributions on ${\cal X}\times{\cal Y}$, let $L$ be an arbitrary loss function on ${\cal Y}\times{\cal A}$, and let $\delta$ be the minimax…