1: \begin{abstract}

2: In this paper we consider  estimation of sparse covariance matrices and

3: propose a thresholding procedure which is adaptive to the variability

4: of individual entries. The estimators are fully data driven and

5: enjoy excellent performance both theoretically and

6: numerically. It is shown that the estimators adaptively achieve the

7: optimal rate of convergence over a large class of sparse covariance

8: matrices under the spectral norm. In

9: contrast, the commonly used universal thresholding estimators are shown to be

10: sub-optimal over the same parameter spaces. Support recovery is also discussed.

11: The adaptive thresholding estimators are easy to implement.

12: Numerical performance of the estimators is studied using both simulated and real data. Simulation results show that the adaptive thresholding estimators uniformly outperform the universal thresholding estimators. The method is also illustrated in an analysis on a dataset from

13: a small round blue-cell tumors microarray experiment. A supplement to this paper which contains additional technical proofs is available online.

14: \end{abstract}

15: