Answers are grounded in this paper and cite their sources.
Let αₙ(·)=P(Xₙ₊₁∈· X₁,…,Xₙ) be the predictive distributions of a sequence (X₁,X₂,…) of p-variate random variables. Suppose αₙ=Nₚ(Mₙ,Qₙ) where Mₙ=1/n∑ᵢ₌₁ⁿXᵢ and Qₙ=1/n∑ᵢ₌₁ⁿ(Xᵢ-Mₙ)(Xᵢ-Mₙ)ᵗ. Then, there is a random probability measure α on Rᵖ such that αₙ→α weakly a.s. If p∈\1,2\, one also obtains αₙ-αa.s. 0 where · is total variation distance. Moreover, the convergence rate of αₙ-α is arbitrarily close to n-1/2. These results (apart from the one regarding the convergence rate) still apply even if αₙ=Lₚ(Mₙ,Qₙ), where Lₚ belongs to a class of distributions much larger than the normal. Finally, the asymptotic behavior of copula-based predictive distributions (introduced in [13]) is investigated and a numerical experiment is performed.
No takes yet. Share an insight, caveat, or question.
Garelli et al. (2024) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: