Thin wrapper around Nestimate::network_reliability() that preserves
the htna actor partition on every model in the returned $models
slot, so downstream htna-aware code (plotting, centralities, etc.)
keeps working on the reliability output.
Arguments
- ...
One or more htna networks built by
build_htna().- iter
Integer. Number of split-half iterations. Default
1000.- split
Numeric in (0, 1). Proportion of sessions used for the first half-network. Default
0.5.- scale
One of
"none","minmax","standardize","proportion". Forwarded toNestimate::network_reliability().- seed
Optional integer seed for reproducibility. Default
NULL(no seed reset).
Value
An object of class c("htna_reliability", "net_reliability"),
with the same components as Nestimate::network_reliability() —
iterations, summary, models, iter, split, scale — and
each entry of $models carrying the htna actor partition
($nodes$groups, $node_groups, $actor_levels).
Details
Mirrors the signature of the underlying function: pass one or more
networks via ... plus the optional iter, split, scale, and
seed arguments. All networks passed through ... must be htna
networks built by build_htna().
Examples
# \donttest{
data(human_ai)
net <- build_htna(human_ai, actor_type = "actor_type")
#> Warning: A network with one long sequence is not recommended and can't be validated using bootstrap and other confirmatory testings.
#> Metadata aggregated per session: ties resolved by first occurrence in 'session_date' (1 sessions), 'cluster' (42 sessions), 'actor_type' (24 sessions)
rel <- reliability_htna(net, iter = 30, seed = 1)
rel$summary
#> model metric mean sd
#> 1 relative mean_dev 0.012347097 0.0009948951
#> 2 relative median_dev 0.007373837 0.0008943957
#> 3 relative cor 0.982368095 0.0041061329
#> 4 relative max_dev 0.097975625 0.0288536573
# }