Answers are grounded in this paper and cite their sources.
Let l(θ) = n⁻¹ log p(x, θ) be the log likelihood of an n-dimensional X under a p-dimensional θ. Let θ̂ⱼ be the mle under Hⱼ: θ¹ = θ¹₀, …, θʲ = θʲ₀ and θ̂₀ be the unrestricted mle. Define Tⱼ as 2n(θ̂j - 1) - l(θ̂ⱼ)\1/2 sgn(θ̂ʲj - 1 - θʲ₀). Let T = (T₁, …, Tₚ). Then under regularity conditions, the following theorem is proved: Under θ = θ₀, T is asymptotically N(n-1/2a₀ + n⁻¹a, J + n⁻¹∑) + O(n-3/2) where J is the identity matrix. The result is proved by first establishing an analogous result when θ is random and then making the prior converge to a degenerate distribution. The existence of the Bartlett correction to order n-3/2 follows from the theorem. We show that an Edgeworth expansion with error O(n⁻²) for T involves only polynomials of degree less than or equal to 3 and hence verify rigorously Lawley's (1956) result giving the order of the error in the Bartlett correction as O(n⁻²).
No takes yet. Share an insight, caveat, or question.
Bickel et al. (1990) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: