Latent Variable 4-Dim Graphical Model Data Example
This data set contains a matrix containing information on four gaussian variables and the corresonding DAG model containing four observed and one latent variable.
data(gmL)
The format is a list of 2 components
$ x: num [1:10000, 1:4] 0.924 -0.189 1.016 0.363 0.497 ... ..- attr(*, "dimnames")=List of 2 .. ..$ : NULL .. ..$ : chr [1:4] "2" "3" "4" "5"
$ g:Formal class 'graphNEL' [package "graph"] with 6 slots .. ..@ nodes : chr [1:5] "1" "2" "3" "4" ... .. ..@ edgeL :List of 5 ........
The data was generated as indicated below. First, a random DAG model was generated with five nodes; then 10000 samples were drawn from this model; finally, variable one was declared to be latent and the corresponding column was deleted from the simulated data set.
## Used to generate "gmL"
set.seed(47)
p <- 5
n <- 10000
gGtrue <- randomDAG(p, prob = 0.3) ## true DAG
myX <- rmvDAG(n, gGtrue)
colnames(myX) <- as.character(1:5)
gmL <- list(x = myX[,-1], g = gGtrue)data(gmL)
str(gmL, max=3)
## the graph:
gmL$g
graph::nodes(gmL$g) ; str(graph::edges(gmL$g))
if(require("Rgraphviz"))
plot(gmL$g, main = "gmL $ g -- latent variable example data")
pairs(gmL $x) # the dataPlease choose more modern alternatives, such as Google Chrome or Mozilla Firefox.