diyano / copulastatistic

All code related to the paper: "A Copula Statistic for Measuring Nonlinear Multivariate Dependence"

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

copulastatistic

All code related to the paper: "A Copula Statistic for Measuring Nonlinear Multivariate Dependence." The authors of this paper are: Mohsen Ben Hassine (mohsenmbh851@gmail.com), Lamine Mili (lmili@vt.edu), and Kiran Karra (kiran.karra@gmail.com).

copulastatistic.R

Provides code to compute the CoS statistic for 2-D,3-D, and 4-D data currently. We will generalize this into one function that can detect the dimensionality and automatically process that data. The original code was provided by Mohsen Ben Hassine.

algorithms.R

Other dependence metric algorithms used to compare against the CoS Note: please use the minerva-r package here: https://github.com/stochasticresearch/minerva-r in order to compute the MICe rather than the stock minerva package from CRAN, which computes the MIC.

codefigX.R

The code required to plot Figure (X) in the paper.

tableX.R

The code required to generate results for Table (X) in the paper.

financialData.R

The code required to generate results relaed to the Financial data in the paper.

yeast.R

The code required to generate the results for Fig.14 in the paper.

ecoli.R

The code required to generate the results for Fig.16 in the paper.

run_reshef_power.R, parse_reshef_power.R

The code required to generate and print the power curves in Fig. 12 of the paper, respectively.

About

All code related to the paper: "A Copula Statistic for Measuring Nonlinear Multivariate Dependence"


Languages

Language:R 100.0%