htna-named wrapper for Nestimate::permutation(). Tests whether
observed edge-weight differences between two networks (or all
pairwise differences within a netobject_group) exceed what would
be expected under a null of identical generating processes.
Arguments
- x, y
The two networks to compare, or a grouped network (
htna_group, which is also anetobject_group) inxwithy = NULLfor all pairwise cohort comparisons. SeeNestimate::permutation().- ...
Additional arguments passed to
Nestimate::permutation(), such asiter,alpha,paired,adjust,measures,nlambda,seed, andactor. See that function for the full, current argument list.
Value
An object of class net_permutation (single pair) or a named
list of class c("net_permutation_group", "list") with one
net_permutation per cohort pair. Both render with
plot_htna_diff(). See Nestimate::permutation() for the full slot
list.
Details
Works on htna networks: the actor partition ($node_groups,
$actor_levels, htna class) is preserved on result$x /
result$y, so plot_htna_diff() can render the result with
htna's colour and layout conventions.
Suffixed _htna to avoid clashing with Nestimate::permutation()
when both packages are loaded.
See also
plot_htna_diff() to plot the result.
Examples
data(human_ai)
grp <- build_htna(human_ai, actor_type = "actor_type", group = "phase")
#> Metadata aggregated per session: ties resolved by first occurrence in 'cluster' (24 sessions), 'actor_type' (9 sessions)
#> Metadata aggregated per session: ties resolved by first occurrence in 'session_date' (1 sessions), 'cluster' (18 sessions), 'actor_type' (15 sessions)
# Pass the grouped network as it is; every cohort pair is tested.
permutation_htna(grp, iter = 50, seed = 1)
#> Grouped Permutation Test
#> Groups: Late vs Early
#>
#> -- Late vs Early --
#> Permutation Test: Transition Network (relative probabilities) [directed]
#> Iterations: 50 | Alpha: 0.05
#> Nodes: 12 | Edges tested: 135 | Significant: 28
#> Global test (networks differ overall?): M = 2.760 (p = 0.0196) | S = 0.181 (p = 0.0196)
