You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Ideally, there would be a way to evaluate some function on each dyad's value and then evaluate the terms of interest on the valued network with new dyad values.
In practice, a good starting point might be a Pow(formula, pow) operator that takes the dyad values to the specified positive power. This could work well with in combination with Project() to model various forms of bipartite mixing while reducing the risk of degeneracy, e.g.,
y ~ Proj1(~Pow(~absdiff("x"), 1/2))
Any other ideas?
The text was updated successfully, but these errors were encountered:
On further thought, using a formula-within-a-formula is more elegant:
y ~ Proj1(~(sqrt~absdiff("x")))
The LHS can have some presets for now, but, eventually, perhaps a stack-based language-within-a-language akin to the one that implements layer logic in ergm.multi could be used.
Ideally, there would be a way to evaluate some function on each dyad's value and then evaluate the terms of interest on the valued network with new dyad values.
In practice, a good starting point might be a
Pow(formula, pow)
operator that takes the dyad values to the specified positive power. This could work well with in combination withProject()
to model various forms of bipartite mixing while reducing the risk of degeneracy, e.g.,y ~ Proj1(~Pow(~absdiff("x"), 1/2))
Any other ideas?
The text was updated successfully, but these errors were encountered: