SNIJDERS, T., and NOWICKI, K. (1997), "Estimation and prediction for stochastic blockmodels for graphs with latent block structure", Journal of Classification, 14, 75 -- 100.  Home/Search   Document Not in Database   Summary   Related Articles

....is much more di#cult when attributes cannot be observed and the class structure can only be identified a posteriori based on the observed relational data y. Wasserman and Anderson (1987) proposed an a posteriori blocking procedure in the framework of the p 1 family (Holland and Leinhardt, 1981) Snijders and Nowicki (1997) studied an a posteriori blockmodel for undirected graphs assuming that C = 2 and that the probability of an edge between two actors depends only on the classes to which the actors belong. For small graphs they proposed ML estimation of the parameters based on the EM algorithm, and for large ....

....of an edge between two actors depends only on the classes to which the actors belong. For small graphs they proposed ML estimation of the parameters based on the EM algorithm, and for large graphs a Bayesian estimation method based on Gibbs sampling. In this paper we extend the approach of Snijders and Nowicki (1997) to the case where relations can be directed and can have an arbirary set of possible values, and where the number of classes is arbitrary. The vector of attributes x specifying the class structure is considered to be unobserved (latent) Conditional on the vector x, we model y using a ....

[Article contains additional citation context not shown here]

SNIJDERS, T., and NOWICKI, K. (1997), "Estimation and prediction for stochastic blockmodels for graphs with latent block structure", Journal of Classification, 14, 75 -- 100.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC