Description Details Author(s) References See Also Examples
Geometrically weighted non-edgewise shared partner distribution for first and second modes of bipartite networks. These terms are closely related to the gwnsp
term in the ergm package, with adaptations to work for bipartite networks and distinguish between nsp's on each mode
gwb1nsp(alpha=0, fixed=TRUE) (binary) (undirected) (curved)
Geometrically weighted nonedgewise shared partner distribution for the first mode in a bipartite (aka two-mode) network: This term is just like gwnsp except it adds a statistic equal to the geometrically weighted nonedgewise (that is, over dyads that do not have an edge) shared partner distribution with weight parameter alpha for the first mode in a bipartite network. The argument fixed indicates whether the scale parameter lambda is to be fit as a curved exponential-family model (see Hunter and Handcock, 2006). Currently, the fixed argument must be set to TRUE, which means the weight parameter alpha is not separately estimated and thus the model is not a CEF model. This term can only be used with undirected bipartite networks.
gwb2nsp(alpha=0, fixed=TRUE) (binary) (undirected) (curved)
Geometrically weighted nonedgewise shared partner distribution for the second mode in a bipartite (aka two-mode) network: This term is just like gwnsp except it adds a statistic equal to the geometrically weighted nonedgewise (that is, over dyads that do not have an edge) shared partner distribution with weight parameter alpha for the second mode in a bipartite network. The argument fixed indicates whether the scale parameter lambda is to be fit as a curved exponential-family model (see Hunter and Handcock, 2006). Currently, the fixed argument must be set to TRUE, which means the weight parameter alpha is not separately estimated and thus the model is not a CEF model. This term can only be used with undirected bipartite networks.
Note that allthough the default is fixed=FALSE
the terms are not yet able to handle a non-fixed decay term so it must be set to TRUE.
Matthew Hamilton matthewlhamilton@gmail.com
Hunter, D. R. and M. S. Handcock (2006). Inference in curved exponential family models for networks. Journal of Computational and Graphical Statistics, 15: 565-583.
Wang, P., Sharpe, K., Robins, G. L., and Pattison, P. E. (2009). Exponential random graph (p*) models for affiliation networks. Social Networks. 31(1): 12-25.
See also gwnsp
term.
1 2 3 4 5 6 7 8 9 10 11 12 | require(ergm.terms.contrib)
# create example data
bel <- data.frame(actor = c("a","a","a","b","b","b","b","c","c","c","c"),
event = c("e1","e2","e3","e3","e1","e4","e5","e3","e1","e4","e6"))
bnet <- network(bel,bipartite=3,directed=FALSE)
# how many actor (first mode) dyads have ANY shared partners?
summary(bnet ~ gwb1nsp(0,TRUE))
# calculate for 2nd mode, using non-extreme alphas
# to weight each additional shared partner less than the one before
summary(bnet ~ gwb2nsp(.5,TRUE))
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.