confirm_edges: Confirm Edges

In GGMncv: Gaussian Graphical Models with Nonconvex Regularization

Description

Confirmatory hypothesis testing of edges that were initially detected with data-driven model selection.

Usage

confirm_edges(object, Rnew, method, alpha)

Arguments

object	An object of class ggmncv
Rnew	Matrix. A correlation matrix of dimensions p by p.
method	Character string. A correction method for multiple comparison (defaults to fdr). Can be abbreviated. See p.adjust.
alpha	Numeric. Significance level (defaults to 0.05).

Details

The basic idea is to merge exploration with confirmation \insertCite@see for example, @rodriguez_williams_rast_mulder_2020GGMncv. This is accomplished by testing those edges (in an independent dataset) that were initially detected via data driven model selection.

Confirmatory hypothesis testing follows the approach described in \insertCitejankova2015confidence;textualGGMncv: (1) graphical lasso is computed with lambda fixed to \mjseqn\lambda = \sqrtlog(p)/n, (2) the de-sparsified estimator is computed, and then (3) p-values are obtained for the de-sparsified estimator.

Value

An object of class ggmncv, including:

• P: Matrix of confirmed edges (partial correlations)

• adj: Matrix of confirmed edges (adjacency)

References

\insertAllCited

Examples

Y <- na.omit(bfi[,1:25])

Y_explore <- Y[1:1000,]

Y_confirm <- Y[1001:nrow(Y),]

fit <- ggmncv(cor(Y_explore),
              n = nrow(Y_explore),
              progress = FALSE)

confirm <- confirm_edges(fit,
                         Rnew = cor(Y_confirm),
                         method = "fdr",
                         alpha = 0.05)

GGMncv documentation built on Dec. 15, 2021, 9:10 a.m.