Estimation of Latent Order Logistic (LOLOG) Models for Networks.
LOLOGs are a flexible and fully general class of statistical graph models.
This package provides functions for performing MOM, GMM and variational
inference. Visual diagnostics and goodness of fit metrics are provided.
See Fellows (2018)
LOLOG is a general framework for generative statistical modeling of graph datasets motivated by the principle of network growth. This class of models is fully general and terms modeling different important network features can be mixed and matched to provide a rich generative description of complex networks.
lologpackage, see the introductory vignette.
To install the latest release from CRAN run:
To install the latest development version from the github repo run:
# If devtools is not installed: # install.packages("devtools") devtools::install_github("statnet/lolog")
If this is your first R source package that you have installed, you’ll also need a set of development tools. On Windows, download and install Rtools, and
devtools takes care of the rest. On a Mac, install the Xcode command line tools. On Linux, install the R development package, usually called
r-base-dev. For details see Package Development Prerequisites.
library(lolog) library(network) data(ukFaculty) # Delete 2 vertices missing group delete.vertices(ukFaculty, which(is.na(ukFaculty %v% "Group"))) # A dyad independent model fitind <- lolog(ukFaculty ~ edges() + nodeMatch("GroupC") + nodeCov("GroupC")) summary(fitind)
This package is set up as an Eclipse project, and the C++ code can be compiled and run without reinstalling the package. To set up in your eclipse IDE, select import project -> General -> Existing Projects into Workspace and select the lolog directory.
This project was set up following the methods outlined in: