compsket

Implementation of the complementary sketching algorithm for two sample testing of high-dimensional regression coefficients.

Description of files

Python code

In the ./python/ folder

• compsket.py: file for the main algorithms
• realdata.py: file for implementing the real data example in Gao and Wang (2020)
• example.ipynb: IPython Notebook for the real data example

R package

In ./R/ and ./man/ folders. Can be installed via devtools::install_github('wangtengyao/compsket') in R.

MATLAB code (with a possible parallel computing implementation)

In the ./matlab/ folder

• complementarySketching.m: function for the main testing algorithm
• differentialNetworkAnalysis.m: specialised function for the nodewise regression testing on the gene interaction network example
• main.m: the script file processing the attached dataset
• CD4_goodTREG_in_thymus.mat: the preprocessed data for Matlab as in ./data/

Data

• CD4_TREG_in_thymus.csv: preprocessed data for the real data example in Section 5 of Gao and Wang (2020).

Reference

Gao, F. and Wang, T. (2020) Two-sample testing of high-dimensional linear regression coefficients via complementary sketching. Preprint, arxiv:2011.13624.