Compute Node Predictability

Computes the proportion of variance explained (R\(^2\)) for each node in the network, following Haslbeck & Waldorp (2018).

For method = "glasso" or "pcor", predictability is computed analytically from the precision matrix: $$R^2_j = 1 - 1 / \Omega_{jj}$$ where \(\Omega\) is the precision (inverse correlation) matrix.

For method = "cor", predictability is the multiple R\(^2\) from regressing each node on its network neighbors (nodes with non-zero edges).

Usage

predictability(object, ...)

# S3 method for class 'netobject'
predictability(object, data = NULL, ...)

# S3 method for class 'netobject_ml'
predictability(object, ...)

# S3 method for class 'netobject_group'
predictability(object, ...)

Arguments

object

A netobject, netobject_ml, or netobject_group object.

...

Additional arguments (ignored).

data

Optional data frame of the original variables used to estimate the network. R\(^2\) never needs it (it comes from the precision or correlation matrix stored on the object). It is used only for the RMSE column and defaults to object$data; when neither is available RMSE is NA.

Value

For netobject: a data frame with one row per node and columns node (character), R2 (numeric, between 0 and 1) and RMSE (numeric, NA when no data is available).

For netobject_ml: a list with elements $between and $within, each such a data frame.

For netobject_group: a named list of such data frames, one per group.

A data frame with one row per node and columns node, R2 and RMSE.

A list with between and within predictability data frames.

A named list of per-group predictability data frames.

References

Haslbeck, J. M. B., & Waldorp, L. J. (2018). How well do network models predict observations? On the importance of predictability in network models. Behavior Research Methods, 50(2), 853–861. doi:10.3758/s13428-017-0910-x

Examples

set.seed(42)
mat <- matrix(rnorm(60), ncol = 4)
colnames(mat) <- LETTERS[1:4]
net <- build_network(as.data.frame(mat), method = "glasso")
predictability(net)
#>   node R2      RMSE
#> 1    A  0 0.9904791
#> 2    B  0 1.3109578
#> 3    C  0 0.9630332
#> 4    D  0 1.0523124