Mixed graph model estimate

MixedGraph is the main function of the package, it uses the BRAIL to do the regression.

MixedGraph(X, crf_structure, family = NULL, rule = c("AND", "OR"),
  brail_control = NULL)

Arguments

X	is a list containing k matrices, each of the matrices has n rows and pk columns. They are the 'blocks'. Each block should have all columns of the same type.
crf_structure	is a K-by-K matrix giving the structure of the CRF element (i, j) is 1 if there are directed edges going from the i-th block to the j-th block; element (i, j) is 0 if there are no edges going from the i-th block to the j-th block; element (i, i) is 1 if there are undirected edges within the i-th block; element (i, i) is 0 if there are not undirected edges within the i-th block
family	is a K vector giving the data types for each blocks. The type could be "gaussian", "binomial" or "poisson". If NULL, we can use a heuristic to guess the right value
rule	says whether to use the "AND" or "OR" rule
brail_control	is a list of optional arguments to be passed to BRAIL internally

Value

the coefficients vector

data

is a list containing k matrices from input

network

is a p * p (sparse) matrix giving the edge weights returned by BRAIL

family

is a K vector giving the data types for each blocks from input, if the input family is NULL, there will be the values guessed in the MixedGraph function

crf_structure

is a K-by-K matrix giving the structure of the CRF from input

stability

is a p * p (sparse) matrix giving the stability scores

rule

"AND" or "OR" rule for edge selection within blocks

Note

The function supports foreach package, so user can register the backend.

Examples

X1 <- matrix(rnorm(12), nrow = 4)
X2 <- matrix(rnorm(12), nrow = 4)
X <- list(X1, X2)
crf_structure <- matrix(c(1, 1, 0, 1), nrow = 2)
brail_control <- list(B = 5, tau = 0.6)
MixedGraph(X, crf_structure, brail_control = brail_control)
#> $data
#> $data[[1]]
#>            [,1]       [,2]      [,3]
#> [1,]  0.2343405 -1.3337463 0.6675088
#> [2,] -0.9515281  1.7647248 1.3728761
#> [3,]  1.8984394  0.5060634 0.5945820
#> [4,] -1.0427705 -0.6552922 0.8093074
#> 
#> $data[[2]]
#>            [,1]        [,2]       [,3]
#> [1,] -0.9293517 -0.11676961 -0.2194639
#> [2,] -0.9688826 -0.04977844 -0.2799220
#> [3,] -0.5634830  0.41253090 -0.1844666
#> [4,]  1.8884887  0.33900622 -1.1310412
#> 
#> 
#> $network
#>            [,1]       [,2]        [,3]      [,4]      [,5]       [,6]
#> [1,]  0.0000000 -0.2888218 -0.19259770  0.000000 0.0000000  0.0000000
#> [2,]  0.0000000  0.0000000  0.15247658  0.000000 0.0000000  0.0000000
#> [3,] -3.0133640  0.0000000  0.00000000  0.000000 0.0000000  0.0000000
#> [4,] -0.5548038 -0.4676904 -0.08437963  0.000000 0.1156798 -0.3241862
#> [5,]  0.0000000  0.0000000  0.00000000  0.000000 0.0000000  0.0000000
#> [6,]  0.0000000  0.0000000 -0.03367474 -2.945617 0.0000000  0.0000000
#> 
#> $family
#> [1] "gaussian" "gaussian"
#> 
#> $crf_structure
#>      [,1] [,2]
#> [1,]    1    0
#> [2,]    1    1
#> 
#> $stability
#>      [,1] [,2] [,3] [,4] [,5] [,6]
#> [1,]  0.0  1.0  1.0  0.0    0  0.0
#> [2,]  0.2  0.0  0.6  0.0    0  0.0
#> [3,]  0.4  0.4  0.0  0.0    0  0.0
#> [4,]  1.0  0.6  1.0  0.0    1  0.6
#> [5,]  0.2  0.2  0.0  0.0    0  0.0
#> [6,]  0.2  0.0  0.8  0.6    0  0.0
#> 
#> $rule
#> [1] "AND"
#> 
#> attr(,"class")
#> [1] "MixedGraph"