Dynamic Slicing for Hypothesis Testing

Desciption

Dynamic slicing is a novel method for detecting interaction between two groups of observations. Specifically, we develop dynamic slicing scheme to deal with non-parametric K-sample test. Dynamic slicing is implemented in C++ and does not require any additional library. But for this program, it requires GNU C library and GNU scientic library (GSL, version: 1.15). Unix/Linux systems are recommended.

User's manual

Click here for user's manual.

Source code and software download

Click here for dynamic slicing source code and software.

Relationship between penalty and Type I error

Click here.

Citation

If you use dynamic slicing, citation should be: Jiang, B., Ye, C. and Liu, J.S. (2013) Non-parametric K-sample Tests via Dynamic Slicing. Preprint.

Contact us

Dr. Bo Jiang
E-mail: bjiang@fas.harvard.edu
Chao Ye
E-mail: yec09@mails.tsinghua.edu.cn
Prof. Jun S. Liu
E-mail: jliu@stat.harvard.edu



Last modified on Feb 14, 2014