CSCS

This implements the CSCS algorithm for covariance and DAG estimation from A convex framework for high-dimensional sparse Cholesky based covariance estimation by Kshitij Khare, Sang Oh, Syed Rahman and Bala Rajaratnam

Basic scripts

The program consists of the following scripts

• data_generate.py: used to generate random multivariate data accoriding to a graph
• CSCS.py: contains the main functions and class for CSCS to estimate DAG and the cholesky parameter for the covariance matrix
• main.py: runs the program to generate the DAG

Notebooks

A notebook with an example is also included

• CSCS_in_python_example.ipynb

Example

import numpy as np
import networkx as nx
from data_generate import generate_random_MVN_data
from CSCS import CSCS

np.random.seed(3689)
Y = generate_random_MVN_data()
cscs = CSCS(Y = Y,l = 1)
L,A,G = cscs.fit()

Authors

• Syed Rahman