The posterior distributions created by perf_mod()
can be used to obtain
the posterior distribution of the difference(s) between models. One or more
comparisons can be computed at the same time.
contrast_models(x, list_1 = NULL, list_2 = NULL, seed = sample.int(10000, 1))
A data frame of the posterior distribution(s) of the difference(s).
The object has an extra class of "posterior_diff"
.
An object produced by perf_mod()
.
Character vectors of equal length that specify the
specific pairwise contrasts. The contrast is parameterized as
list_1[i] - list_2[i]
. If the defaults are left to NULL
, all
combinations are evaluated.
A single integer for sampling from the posterior.
If a transformation was used when x
was created, the inverse is
applied before the difference is computed.