Compute Statistics

Computes observed sufficient statistics for the terms specified in a formula.

Usage

statistics(formula, canonical_names = FALSE)

Arguments

formula

A model `formula` object. The left-hand side should be the name of a iglm.data object available in the calling environment. Alternatively, the left-hand side can be a iglm.data.list object to compute statistics for multiple iglm.data objects at once (e.g., the output of simulate_iglm). See iglm.terms for details on specifying the right-hand side terms.

canonical_names

(logical) If `TRUE`, returns canonical term names without label substitution. Default is `FALSE`.

Value

A named numeric vector (or matrix if a list of data objects is provided). Each element corresponds to a term in the `formula`, and its value is the calculated observed feature for that term based on the data in the iglm.data object. The names of the vector match the (optionally labeled) term names derived from the formula terms.

See also

iglm, iglm.terms, iglm.data

Examples

# Create an iglm.data object
n_units <- 10
neighborhood <- matrix(1, nrow = n_units, ncol = n_units)
family_x <- "binomial"
family_y <- "binomial"
x_attr_data <- rbinom(n_units, 1, 0.5)
y_attr_data <- rbinom(n_units, 1, 0.5)
z_net_data <- matrix(0, nrow = n_units, ncol = n_units)
object <- iglm.data(
  z_network = z_net_data, x_attribute = x_attr_data,
  y_attribute = y_attr_data, neighborhood = neighborhood,
  directed = FALSE, family_x = family_x, family_y = family_y,
  n_units = n_units,
  label_x = "age", label_y = "income"
)
statistics(object ~ edges(mode = "local") + attribute_y + attribute_x)
#> edges(mode = 'local')      attribute_income         attribute_age 
#>                     0                     4                     4 
statistics(object ~ edges(mode = "local") + attribute_y + attribute_x, canonical_names = TRUE)
#> edges(mode = 'local')           attribute_y           attribute_x 
#>                     0                     4                     4