bdgraph.sim: Synthetic graph data generator

Description

Implements a synthetic graph data generation with different graph structures, including "random", "cluster", "scale-free", "hub" "fixed", and "circle". The default is "random". Based on the graph structure, it generate four differents data sets which is multivariate Gaussian, non-Gaussian, discrete, or mixed data.

Usage

bdgraph.sim( n = 2, p = 10, graph = "random", size = NULL, prob = 0.2, 
             class = NULL, type = "Gaussian", cut = 4, b = 3, D = diag(p), 
             K = NULL, sigma = NULL, mean = 0, vis = FALSE )

Arguments

The number of samples required. The default value is 2.

The number of variables (nodes). The default value is 10.

graph

The graph structure with option "random" (as a default), "cluster", "scale-free", "hub" "fixed", and "circle". It also could be adjacency matrix corresponding to the graph structure (an upper triangular matrix in which $g_{ij}=1$ if there is

size

The number of links in true graph (graph size).

prob

For "random" graph, it is the probability that a pair of nodes has a link. The default value is 0.2.

class

Is number of the classes for "cluster" option.

type

Type of data which has four options "Gaussian" (as a default), "non-Gaussian", "discrete", and "mixed". For option "Gaussian", data is generated from multivariate normal distribution. For option "non-Gaussian", data is transfered multivariate normal

cut

For simulated discrete data (type = "discrete"), it is number of categories for discrete variables. The default value is 4.

The degree of freedom for G-Wishart distribution, $W_G(b, D)$. The default is 3.

The positive definite $(p \times p)$ "scale" matrix for G-Wishart distribution, $W_G(b, D)$. The default is an identity matrix.

A positive-definite symmetric matrix specifying the precision matrix of the variables. It is for "fixed" option of graph.

sigma

A positive-definite symmetric matrix specifying the covariance matrix of the variables. It is for "fixed" option of graph.

mean

A vector specifying the mean of the variables. The default value is zero.

vis

visualize the true graph pattern. The default value is FALSE.

Value

an object with S3 class "simulate" as follow:
dataGenerated data which is a $n \times p$ matrix.
sigmaThe covariance matrix of the generated data.
KThe precision matrix of the generated data.
GThe adjacency matrix corresponding to the true graph structure.

References

Mohammadi, A. and Wit, E. C. (2014). Bayesian structure learning in sparse Gaussian graphical models, Bayesian Analysis, acceped. http://arxiv.org/abs/1210.5371v6

Examples

Run this code

# generating synthetic multivariate normal data from a 'random' graph
		data.sim <- bdgraph.sim( n = 50, p = 10, prob = 0.3, vis = TRUE )
		print(data.sim)

		# generating synthetic multivariate normal data from a 'hub' graph
		data.sim <- bdgraph.sim( n = 3, p = 6, graph = "hub", vis = FALSE )
		data.sim $ data

		# generating synthetic mixed data from a 'hub' graph 
		data.sim <- bdgraph.sim( n = 10, p = 8, graph = "hub", type = "mixed" )
		data.sim $ data