1: \begin{abstract}

2: \textbf{Abstract}. Many popular statistical models, such as factor

3: and random effects models, give arise a certain type of covariance

4: structures that is a summation of low rank and sparse matrices. This

5: paper introduces a penalized approximation framework to recover such

6: model structures from large covariance matrix estimation. We propose

7: an estimator based on minimizing a non-likelihood loss with separable

8: non-smooth penalty functions. This estimator is shown to recover exactly

9: the rank and sparsity patterns of these two components, and thus partially

10: recovers the model structures. Convergence rates under various matrix

11: norms are also presented. To compute this estimator, we further develop

12: a first-order iterative algorithm to solve a convex optimization problem

13: that contains separable non-smooth functions, and the algorithm is

14: shown to produce a solution within $O(t^{-2})$ of the optimal, after

15: any finite $t$ iterations. Numerical performance is illustrated using

16: simulated data and stock portfolio selection on S\&P 100.

17: \end{abstract}

18: