Simple null

Theorem: If regularity conditions (A)-(D) hold, then

(θ^θ)In(θ)(θ^θ)dχd2.

This follows directly from the asymptotic normality of the MLE.

Nuisance parameters

Theorem: If regularity conditions (A)-(D) hold and θ0=θ1 (i.e., if H0 is true), then

n(θ^1θ0)dN(0,V11),

where V111=I11I12I221I21 is the portion of I(θ) corresponding to θ1.

This follows directly from the asymptotic normality of the MLE and the marginal distribution of a multivariate normal.

Confidence intervals

If our parameter of interest is a scalar, then we have simple closed-form expressions for confidence intervals:

θ^j±z1α/2Vjjn(θ^).