Visible to Intel only — GUID: GUID-09291952-1579-4A3D-A1FC-5FF32B4901D0
Visible to Intel only — GUID: GUID-09291952-1579-4A3D-A1FC-5FF32B4901D0
Basic Assumptions for the MI Method
The MI method is provided under the following assumptions:
The base model for the Summary Statistics version of MI is a multivariate normal distribution with parameters (μ, Σ) where
μ is a vector of means.
Σ is a variance-covariance matrix.
Prior distribution of μ is a conditionally-multivariate Gaussian given Σ with parameters μ0∈R7 and τ-1Σ, where τ is a positive constant. The variance-covariance matrix Σ follows the inverted-Wishart distribution for fixed parameters m ≥ p and a positive-definite matrix Λ.
Data points are Missed At Random (MAR).
The strict definition of this and other mechanisms supporting missing values are available in [Rubin1987].