Compute Summary Statistics for Scenario Construction

Extracts summary statistics (mean, quantiles) from a 4D covariate array. These values are used to set baseline and varying levels when building counterfactual scenario arrays with get_scen_array.

Usage

get_scen_vals(data, vars = NULL, time = NULL, directed = NULL)

Arguments

data

Four-dimensional array (m x m x p x T) of covariates, or a three-dimensional array (m x m x T) treated as a single variable.

vars

Character vector of variable names (matching the third dimension of data) to summarise. If NULL, all variables are used.

time

Integer vector of time indices to include. If NULL, all time periods are used.

directed

Logical vector of length length(vars) indicating whether each variable is directed. For undirected variables the lower triangle is excluded before computing statistics. Defaults to TRUE for all variables.

Value

A named list. Each element corresponds to one variable and contains:

mean

Scalar mean of off-diagonal entries.

quantiles

Named numeric vector of quantiles at 5% increments from 0 to 1.

Examples

if (FALSE) { # \dontrun{
# W is m x m x p (or m x m x p x T)
vals <- get_scen_vals(W)
vals[["proximity"]]$mean
vals[["proximity"]]$quantiles
} # }