# convertGraph: Convert graphs into adjacency matrices In GGMselect: Gaussian Graphs Models Selection

## Description

Convert into adjacency matrices `NG` graphs (expressed as lists of connected nodes)

## Usage

 `1` ```convertGraph(Graph) ```

## Arguments

 `Graph` array of dimension `p x Dmax x NG`, where `Dmax` is the degree of the graph and `NG` the number of graphs. If `NG` is equal to 1, `Graph` can be a matrix of dimension `p x Dmax`. `Graph[a,,iG]` should be the indices of the nodes connected to the node `a`, for the graph `iG`; `Graph[a,1,iG]` should be equal to 0 if there is no node connected to the node `a`.

## Value

An array of dimension `p x p x NG`, or, when `NG` is equal to 1, a matrix of dimension `p x p`.

The entry `[,,iG]` is a symmetric matrix, with diagonal equal to zero. The entry `[a,b,iG]` is equal to 1 if `a` is connected to `b`, 0 otherwise.

## Note

This function is useful to generate the entry `MyFamily` of the function `selectMyFam`. Actually, the list of adjacency matrices `MyFamily` can be generated from lists of connected nodes with `convertGraph`.

## Author(s)

Bouvier A, Giraud C, Huet S, Verzelen N

## References

Please use `citation("GGMselect")`

`selectQE`, `selectMyFam`, `selectFast`, `simulateGraph`, `penalty`

## Examples

 ``` 1 2 3 4 5 6 7 8 9 10 11 12 13 14``` ```p=30 n=30 # simulate graph eta=0.11 Gr <- simulateGraph(p,eta) X <- rmvnorm(n, mean=rep(0,p), sigma=Gr\$C) # estimate graph GRest <- selectFast(X, family="C01") # Neighb and G are 2 forms of the same result a <- convertGraph(GRest\$C01\$Neighb) print(all.equal(a, GRest\$C01\$G)) # TRUE # recalculate the graph with selectMyFam GMF <- selectMyFam(X, list(a)) print(all.equal(a,GMF\$G)) # TRUE ```

### Example output

```Loading required package: mvtnorm