Robocov_precision: Robocov partial correlation estimation using robust...

In kkdey/Robocov: Robust Estimation of covariance matrix and conditional graphs

Description

A robust estimation of partial correlation matrix for data with missing entries using a robust optimization version of the GLASSO method taking account of the missing entries in the data matrix.

Usage

Robocov_precision(data_with_missing, alpha, lambda = 1)

Arguments

data_with_missing	The samples by features data matrix. May contain NA values.
alpha	The tuning parameter for L-1 penalty.
lambda	The weight of the constraint in the shrinkage. Default lambda taken to be 1.

Examples

data("sample_by_feature_data")
out = Robocov_precision(sample_by_feature_data, alpha = 0.1)
corrplot::corrplot(as.matrix(out), diag = FALSE,
        col = colorRampPalette(c("blue", "white", "red"))(200),
        tl.pos = "td", tl.cex = 0.4, tl.col = "black",
        rect.col = "white",na.label.col = "white",
        method = "color", type = "upper")

kkdey/Robocov documentation built on June 12, 2020, 11:34 a.m.