ergmx.NetSeries#

ergmx.NetSeries(*networks, times=None, bipartite=None, na_impute=None)[source]#

A series of networks on the same vertices, as R’s tergm::NetSeries(), to model each transition conditional on the network before it.

Use it with tergm’s operators: Form() (the ties that formed: terms of the union of the previous and the current network), Persist() and Diss() (the ties that persisted: the intersection), Cross() (the current network) and Change() (the dyads that changed). Their linear models can use .Time (the time of the current network), .TimeID (1 for the first transition, 2…) and .TimeDelta (the time since the previous network). ergmx.tergm() builds the series itself.

Parameters:

  • *networks (igraph.Graph or networkx.Graph) – The networks, in time order, as arguments or a list. They must have the same vertices, in the same order.

  • times (list of numbers, optional) – The times the networks were observed: 0, 1, 2… by default.

  • bipartite (str or bool, optional) – For bipartite networks, the vertex attribute with each vertex’s mode.

  • na_impute (str or list of str, optional) – How to impute the missing dyads of the networks transitioned from (all but the last), as tergm’s NA.impute: "next" (the dyad’s value in the next network), "previous" (in the previous one), "majority" (the more common value among the network’s observed dyads), "0" or "1"; several are applied in turn. The networks transitioned to keep their missing dyads, which the fit treats as missing. Without it, missing dyads in the networks transitioned from are an error.