Skip to content

Commit 486f5a4

Browse files
authored
Update README.md
1 parent d9024eb commit 486f5a4

File tree

1 file changed

+2
-0
lines changed

1 file changed

+2
-0
lines changed

README.md

Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -1,4 +1,6 @@
11
# A-finite-sample-estimator-for-large-covariance-matrices
22
This repository contains data and code relative to the manuscript "A finite sample estimator for large covariance matrices" by Matteo Farnè and Angela Montanari (https://arxiv.org/abs/1711.08950).
3+
34
The MATLAB dataset 'supervisory_data.m' contains the covariance matrix and the labels of a selection of Euro Area banking supervisory data. Punctual data are not available due to confidenatiality reasons. The dataset contains the covariance matrix 'C', and the relative labels of supervisory indicators, 'Labgood'. The labels may be interpreted exploiting the detailed description at the link https://www.eba.europa.eu/documents/10180/359626/Annex+III+-+FINREP+templates+IFRS.xlsx/049e48a4-e7c2-44c6-89b1-4086447bced9
5+
46
Two MATLAB functions, called "UNALCE.m" and "POET.m", are provided. The former performs the new procedure for covariance matrix estimation described in https://arxiv.org/abs/1711.08950, called UNALCE (UNshrunk ALgebraic Covariance Estimator). The latter performs POET covariance estimation procedure (Fan et al., 2013, http://onlinelibrary.wiley.com/doi/10.1111/rssb.12016/abstract). Both functions contain the detailed explanation of input and output arguments.

0 commit comments

Comments
 (0)