Hyperedge analogue of endogenous_features(). Accepts a
hyperedge log (see hyperedge_log()) and computes hyperedge-native
statistics, falling back to the dyadic engine for stat names that
belong to the standard dyadic endogenous catalogue.
Arguments
- hyperedge_log
A hyperedge log (see
hyperedge_log()).- stats
Character vector of statistic names. Mix of hyperedge- native names listed above and the dyadic catalogue names accepted by
endogenous_features().- half_life
Required when an exp-decay statistic is requested (only applies to delegated dyadic stats; hyperedge subrep does not use a half-life).
Details
Recognised hyperedge stat names:
"subrep_<rho>_<l>"Directed subset repetition (paper eq. 4).
rho= sender-side subset cardinality (1..|I|),l= receiver-side subset cardinality (0..|J|, 0 = ignore receivers). Examples:"subrep_1_1"(average activity over single-actor sub-pairs),"subrep_2_1"(over pair-of-senders × single-receiver subpairs)."subrep_<rho>"Undirected subset repetition. Equivalent to
"subrep_<rho>_0"; counts past events whose participant set is a superset of the chosen subset, with no receiver-side restriction."activity"Counts past events whose participant set covers the focal event's entire
(I, J)pair. Equivalent to"subrep_<|I|>_<|J|>".
For dyadic-shaped events (every row has |I| = |J| = 1) and a
dyadic stat name, this function delegates to
endogenous_features() via as_dyadic_log().
References
Boschi M, Lerner J, Wit EC (2025). Beyond Linearity and Time-Homogeneity: Relational Hyper Event Models with Time-Varying Non-Linear Effects. arXiv:2509.05289.
Examples
hl <- hyperedge_log(
I = list(c("a","b"), c("a","c"), c("b","c"), c("a","b","c")),
J = list(c("X"), c("X","Y"), c("Y"), c("X")),
time = c(1, 2, 3, 4))
hyperedge_features(hl,
stats = c("subrep_1_1", "subrep_2_1", "activity"))
#> I J time subrep_1_1 subrep_2_1 activity
#> 1 a, b X 1 0.000000 0.0000000 0
#> 2 a, c X, Y 2 0.250000 0.0000000 0
#> 3 b, c Y 3 0.500000 0.0000000 0
#> 4 a, b, c X 4 1.333333 0.6666667 0