Reports how well each node is predicted by the others in a fitted network.
For Gaussian graphical models this is the closed-form variance explained
(R-squared) from the precision matrix and needs no data. For the nodewise
models (ising_fit(), ising_sampler(), mgm_fit()) it requires the data
and reports R-squared for Gaussian nodes and classification accuracy (CC)
plus normalized accuracy (nCC) for binary nodes.
Prediction selects the original modeled columns, independently of display
labels. Binary MGM categories use the levels recorded when fitting,
including numeric codes such as 1/2; unseen categories are refused.
Arguments
- x
A psychnet object.
- data
The data the network was estimated from; required for the nodewise models (ising / IsingSampler / mgm), ignored for the GGMs.
- ...
Unused.
Value
A tidy data.frame, one row per node, with columns node, type
("gaussian" or "binary"), metric ("R2" or "nCC"),
predictability, and accuracy (classification accuracy for binary nodes,
NA for Gaussian).
Examples
S <- 0.4^abs(outer(1:6, 1:6, "-"))
net_predict(ebic_glasso(cor_matrix = S, n = 250))
#> node type metric predictability accuracy
#> 1 V1 gaussian R2 0.1568160 NA
#> 2 V2 gaussian R2 0.2711166 NA
#> 3 V3 gaussian R2 0.2711166 NA
#> 4 V4 gaussian R2 0.2711166 NA
#> 5 V5 gaussian R2 0.2711166 NA
#> 6 V6 gaussian R2 0.1568160 NA