igraph (version 0.6-3)

revolver: Measuring the driving force in evolving networks

Description

These functions assume a simple evolving network model and measure the functional form of a so-called attractiveness function governing the evolution of the network.

Usage

evolver.d (nodes, kernel, outseq = NULL, outdist = NULL, m = 1, 
           directed = TRUE)

revolver.d (graph, niter=5, sd=FALSE, norm=FALSE, cites=FALSE, expected=FALSE, error=TRUE, debug=numeric()) revolver.ad (graph, niter=5, agebins=max(vcount(graph)/7100, 10), sd=FALSE, norm=FALSE, cites=FALSE, expected=FALSE, error=TRUE, debug=matrix(ncol=2, nrow=0)) revolver.ade (graph, cats, niter=5, agebins=max(vcount(graph)/7100, 10), sd=FALSE, norm=FALSE, cites=FALSE, expected=FALSE, error=TRUE, debug=matrix(ncol=2, nrow=0)) revolver.e (graph, cats, niter=5, st=FALSE, sd=FALSE, norm=FALSE, cites=FALSE, expected=FALSE, error=TRUE, debug=numeric()) revolver.de (graph, cats, niter=5, sd=FALSE, norm=FALSE, cites=FALSE, expected=FALSE, error=TRUE, debug=numeric()) revolver.l (graph, niter=5, agebins=max(vcount(graph)/7100, 10), sd=FALSE, norm=FALSE, cites=FALSE, expected=FALSE, error=TRUE, debug=numeric()) revolver.dl (graph, niter=5, agebins=max(vcount(graph)/7100, 10), sd=FALSE, norm=FALSE, cites=FALSE, expected=FALSE, error=TRUE, debug=numeric()) revolver.el (graph, cats, niter=5, agebins=max(vcount(graph)/7100, 10), sd=FALSE, norm=FALSE, cites=FALSE, expected=FALSE, error=TRUE, debug=numeric()) revolver.r (graph, window, niter=5, sd=FALSE, norm=FALSE, cites=FALSE, expected=FALSE, error=TRUE, debug=numeric()) revolver.ar (graph, window, niter=5, agebins=max(vcount(graph)/7100, 10), sd=FALSE, norm=FALSE, cites=FALSE, expected=FALSE, error=TRUE, debug=matrix(ncol=2, nrow=0)) revolver.di (graph, cats, niter=5, sd=FALSE, norm=FALSE, cites=FALSE, expected=FALSE, error=TRUE, debug=numeric()) revolver.adi (graph, cats, niter=5, agebins=max(vcount(graph)/7100, 10), sd=FALSE, norm=FALSE, cites=FALSE, expected=FALSE, error=TRUE, debug=matrix(ncol=2, nrow=0)) revolver.il (graph, cats, niter=5, agebins=max(vcount(graph)/7100, 10), sd=FALSE, norm=FALSE, cites=FALSE, expected=FALSE, error=TRUE, debug=numeric()) revolver.ir (graph, cats, window, niter=5, sd=FALSE, norm=FALSE, cites=FALSE, expected=FALSE, error=TRUE, debug=numeric()) revolver.air (graph, cats, window, niter=5, agebins=max(vcount(graph)/7100, 10), sd=FALSE, norm=FALSE, cites=FALSE, expected=FALSE, error=TRUE, debug=matrix(ncol=2, nrow=0)) revolver.d.d (graph, vtime = V(graph)$time, etime = E(graph)$time, niter = 5, sd = FALSE, norm = FALSE, cites = FALSE, expected = FALSE, error = TRUE, debug = matrix(ncol = 2, nrow = 0)) revolver.p.p (graph, events = get.graph.attribute(graph, "events"), vtime = V(graph)$time, etime = E(graph)$time, niter = 5, sd = FALSE, norm = FALSE, cites = FALSE, expected = FALSE, error = TRUE, debug = matrix(ncol = 2, nrow = 0)) revolver.error.d (graph, kernel) revolver.error.ad (graph, kernel) revolver.error.ade (graph, kernel, cats) revolver.error.adi (graph, kernel, cats) revolver.error.air (graph, kernel, cats, window) revolver.error.ar (graph, kernel, window) revolver.error.de (graph, kernel, cats) revolver.error.di (graph, kernel, cats) revolver.error.dl (graph, kernel) revolver.error.e (graph, kernel, cats) revolver.error.el (graph, kernel, cats) revolver.error.il (graph, kernel, cats) revolver.error.ir (graph, kernel, cats, window) revolver.error.l (graph, kernel) revolver.error.r (graph, kernel, window)

revolver.ml.ade (graph, niter, cats, agebins = 300, delta = 1e-10, filter = NULL) revolver.ml.d (graph, niter, delta = 1e-10, filter = NULL) revolver.ml.de (graph, niter, cats, delta = 1e-10, filter = NULL) revolver.ml.df (graph, niter, delta = 1e-10) revolver.ml.f (graph, niter, delta = 1e-10) revolver.ml.l (graph, niter, agebins = 300, delta = 1e-10)

revolver.ml.AD.alpha.a.beta (graph, alpha, a, beta, abstol = 1e-08, reltol = 1e-08, maxit = 1000, agebins = 300, filter = NULL) revolver.ml.AD.dpareto (graph, alpha, a, paralpha, parbeta, parscale, abstol = 1e-08, reltol = 1e-08, maxit = 1000, agebins = 300, filter = NULL) revolver.ml.ADE.alpha.a.beta (graph, cats, alpha, a, beta, coeffs, abstol = 1e-08, reltol = 1e-08, maxit = 1000, agebins = 300, filter = NULL) revolver.ml.ADE.dpareto (graph, cats, alpha, a, paralpha, parbeta, parscale, coeffs, abstol = 1e-08, reltol = 1e-08, maxit = 1000, agebins = 300, filter = NULL) revolver.ml.D.alpha (graph, alpha, abstol = 1e-08, reltol = 1e-08, maxit = 1000, filter = NULL) revolver.ml.D.alpha.a (graph, alpha, a, abstol = 1e-08, reltol = 1e-08, maxit = 1000, filter = NULL) revolver.ml.DE.alpha.a (graph, cats, alpha, a, coeffs, abstol = 1e-08, reltol = 1e-08, maxit = 1000, filter = NULL)

revolver.ml.AD.dpareto.eval (graph, alpha, a, paralpha, parbeta, parscale, agebins = 300, filter = NULL) revolver.ml.ADE.dpareto.eval (graph, cats, alpha, a, paralpha, parbeta, parscale, coeffs, agebins = 300, filter = NULL) revolver.ml.ADE.dpareto.evalf (graph, cats, par, agebins, filter = NULL)

revolver.probs.ad (graph, kernel, ntk = FALSE) revolver.probs.ade (graph, kernel, cats) revolver.probs.d (graph, kernel, ntk = FALSE) revolver.probs.de (graph, kernel, cats) revolver.probs.ADE.dpareto (graph, par, cats, gcats, agebins)

Arguments

nodes
The number of vertices in the generated network.
kernel
The kernel function, a vector, matrix or array, depending on the number of model parameters.
outseq
The out-degree sequence, or NULL if no out-degree sequence is used.
outdist
The out-degree distribution, or NULL if all vertices have the same out-degree. This argument is ignored if the outseq argument is not NULL.
m
Numeric scalar, the out-degree of the verticec. It is ignored if at least one of outseq and outdist is not NULL.
directed
Logical scalar, whether to create a directed graph.
graph
The input graph.
niter
The number of iterations to perform.
sd
Logical scalar, whether to return the standard deviation of the estimates.
norm
Logical scalar, whether to return the normalizing factors.
cites
Logical scalar, whether to return the number of citations to the different vertex types.
expected
Logical scalar, whether to return the expected number of citations for the different vertex types.
error
Logical scalar, whether to return the error of the fit.
debug
Currently not used.
agebins
The number of bins for vertex age.
cats
The number of categories to use.
window
The width of the time window to use, measured in number of vertices.
vtime
Numeric vector, the time steps when the vertices where added to the network.
etime
Numeric vector, the time steps when the edges where added to the network.
events
A list of numeric vectors, each vector represents an event, with the participation of the listed vertices.
delta
Real scalar, the error margin that is allowed for the convergence.
filter
Logical vector, length is the number of vertices. Only vertices corresponding to TRUE entries are used in the fitting.
alpha
Starting value for the alpha parameter.
a
Starting value for the a parameter.
paralpha
Starting value for the paralpha (Pareto alpha) parameter.
parbeta
Starting value for the parbeta (Pareto beta) parameter.
parscale
Starting value for the parscale (Pareto scale) parameter.
abstol
Real scalar, absolute tolerance for the ML fitting.
reltol
Real scalar, relative tolerance for the ML fitting.
maxit
Numeric scalar, the maximum number of iterations.
beta
Real scalar, starting value for the beta parameter.
coeffs
Numeric vector, starting values for the coefficients.
par
Pareto parameters for the different vertex types, in a matrix.
ntk
Logical scalar, whether to return the Ntk values.
gcats
Numeric vector, the vertex types.
st
Logical scalar, whether to return the S(t) values.

Value

  • A named list.

Details

The functions should be considered as experimental, so no detailed documentation yet. Sorry.