Coerce an inferential comparison to a network difference
Source:R/subtract_networks.R
as_netdifference.Rdas_netdifference() turns the result of a network comparison into a
plain netdifference object, the signed difference network that
subtract_networks() returns, so that cograph's difference renderer
draws it. It reads a bayes_compare() result (net_bayes), and returns
a netdifference unchanged. The drawn weights are the posterior mean
differences, by default only those of the credible edges, while the full
difference, its credible intervals and its probabilities of direction
are kept beside them.
Usage
as_netdifference(x, ...)
# S3 method for class 'net_bayes'
as_netdifference(x, significant_only = TRUE, ...)
# S3 method for class 'netdifference'
as_netdifference(x, ...)
# Default S3 method
as_netdifference(x, ...)Arguments
- x
The comparison: a
net_bayesfrombayes_compare(), or anetdifference, which is returned as it is. Any other object raises an error.- ...
Not used.
- significant_only
For a
net_bayes: whether the drawn weights keep only the credible differences (thosebayes_compare()marks insig), with every other edge set to 0. The full difference and interval matrices are kept either way. DefaultTRUE.
Value
A netdifference object, which is also a netobject (method
"difference"); print() lists its edges and cograph::splot() draws
it. From a net_bayes, its weights are the posterior mean
differences (all of them, or only the credible ones, by
significant_only), and it keeps the full posterior mean difference
(difference_matrix), the two posterior mean matrices (x, y), the
interval bounds (ci_lower, ci_upper), the Bayesian p-equivalents
(p_values), the probabilities of direction (p_difference), the
credibility matrix (sig), significant_only, and
netdifference_type = "bayes_compare".
Examples
s1 <- data.frame(V1 = c("A", "B", "C"), V2 = c("B", "C", "A"))
s2 <- data.frame(V1 = c("A", "C", "B"), V2 = c("C", "B", "A"))
b <- bayes_compare(build_network(s1, method = "relative"),
build_network(s2, method = "relative"),
draws = 500, seed = 1)
as_netdifference(b, significant_only = FALSE)
#> Network difference (x - y): 3 nodes, 6 differing edges
#> Plot: cograph::splot(d) or cograph::plot_difference(d)
#>
#> from to x y difference
#> B A 0.2 0.6 -0.4
#> C A 0.6 0.2 0.4
#> A B 0.6 0.2 0.4
#> C B 0.2 0.6 -0.4
#> A C 0.2 0.6 -0.4
#> B C 0.6 0.2 0.4
#> A A 0.2 0.2 0.0
#> B B 0.2 0.2 0.0
#> C C 0.2 0.2 0.0