Skip to main content
. Author manuscript; available in PMC: 2015 Jan 2.
Published in final edited form as: Electron J Stat. 2012 Nov 9;6:2125–2149. doi: 10.1214/12-EJS740

Algorithm 3 DP-GLASSO algorithm
  1. Initialize Θ = diag(S + λI)−1.

  2. Cycle around the columns repeatedly, performing the following steps till convergence:
    1. Rearrange the rows/columns so that the target column is last (implicitly).
    2. Solve (5.3) for γ~ and update
      θ^12=Θ11(s12+γ~)/w22
    3. Solve for θ22 using (5.5).
    4. Update the working covariance w12=s12+γ~.