Skip to main content
. Author manuscript; available in PMC: 2016 Feb 11.
Published in final edited form as: Ann Appl Stat. 2010 Jun;4(2):764–790. doi: 10.1214/09-AOAS314

Algorithm 1.

Imputation with Transposable Regularized Covariance Models (TRCMimpute)

  1. Initialization:
    1. Estimate ν̂ and μ̂ from the observed data.
    2. If xij is missing, set xij = ν̂i + μ̂j.
    3. Start with non-singular estimates Σ̂ and Δ̂.

  2. E Step (Δ): Calculate X̂T Σ̂−1X̂ + G(Σ̂−1).

  3. M Step (Δ):
    1. Update estimates of ν̂ and μ̂.
    2. Maximize Q with respect to Δ−1 to obtain Δ̂.

  4. E Step (Σ): Calculate X̂Δ̂−1X̂T + F(Δ̂−1),

  5. M Step (Σ):
    1. Update estimates of ν̂ and μ̂.
    2. Maximize Q with respect to Σ−1 to obtain Σ̂.

  6. Repeat Steps 2–5 until convergence.