# penalty: Penalty function In GGMselect: Gaussian Graphs Models Selection

## Description

Compute the penalty function of GGMselect.

## Usage

 `1` ```penalty(p,n, dmax=min(3,n-3,p-1), K=2.5) ```

## Arguments

 `p` the number of variables. `p` should be greater than `1`. `n` the sample size. `n` should be greater than `3`. `dmax` integer or `p`-dimensional vector of integers smaller or equal to `min(n-3, p-1)`. When `dmax` is a scalar, it gives the maximum degree of the estimated graph. When `dmax` is a vector, `dmax[a]` gives the maximum degree of the node `a`. Default value: `min(3,n-3,p-1)`. `K` scalar or vector of real numbers larger than 1. Tuning parameter of the penalty function.

## Details

More details are available on ../doc/Notice.pdf

## Value

A matrix of dimension `(max(Dmax)+1) x length(K)`. The entry `[d+1,k]` gives the value of the penalty for the dimension `d` and the parameter `K[k]`.

## Author(s)

Bouvier A, Giraud C, Huet S, Verzelen N

## References

Please use `citation("GGMselect")`

`selectQE`, `selectMyFam`, `selectFast`, `simulateGraph`, `convertGraph`

## Examples

 ```1 2 3``` ```p=30 n=30 pen <- penalty(p,n, 3) ```

### Example output

```Loading required package: mvtnorm