*****  To join INSNA, visit  *****

Dear list members,


I currently analyze directed networks of knowledge exchange in 5 different
teams with ERGM. I only worked with PNet up to now, but need to reproduce my
analyses in Statnet/R as the attribute-reciprocity parameters I need are
defined slightly differently in PNet as they are in R and do not fit
completely my theoretical concept. However, I encounter a problem in Statnet
when including the corresponding "mutual" parameter in combination with
quantitative vertex attributes. The term I use is mutual(by="name of the
attribute"). The full model specification is the following:

nwkoopmodel <-
constraints =~ fixedas(absent = elkoopszrev),
control=control.ergm(MCMC.interval = 8000,MCMC.burnin = 8000,SA.phase1_n =
10000,SA.niterations = 30,SA.phase3_n = 10000),verbose=TRUE)

.where "nwkoop" is the name of my network dataset with 138 actors, QA1 and
QA2 are quantitative vertex attributes, BA is a binary vertex attribute. I
also included the constraint term "fixedas" together with an edgelist
("elkoopszrev") defining ties that cannot exist in the network
(corresponding to "structural zeros" in PNet). (Indeed, as some of the teams
I analyze are very small, I decided to integrate them all together in one
"network" for the ERGM. However, ties occur only between members of the same
team, not between teams.)


The model begins to run, but produces an error shortly after indicating "R
encountered a fatal error. The session was terminated.". This problem occurs
with different quantitative attributes. When I run the same model including
"mutual" without the vertex attribute - for standard mutual statistics - it
works fine. Does anyone know if there is something wrong with my syntax or
what else could be the problem?



Any help would be highly appreciated!




Natalie David





Natalie David

Research Associate


Chair for Organisation and HRM

Albert-Ludwigs-University of Freiburg

Platz der Alten Synagoge 1

79085 Freiburg, Germany

Phone: +49 761 203-2392

Fax: +49 761 203-2394

E-Mail:  <mailto:[log in to unmask]>
[log in to unmask]







SOCNET is a service of INSNA, the professional association for social
network researchers ( To unsubscribe, send
an email message to [log in to unmask] containing the line
UNSUBSCRIBE SOCNET in the body of the message.