# Generate a block diagonal covariance matrix

### Description

This function is included in the package so that it can be used in the
example code provided in `spcov`

.

### Usage

1 | ```
GenerateCliquesCovariance(ncliques, cliquesize, theta)
``` |

### Arguments

`ncliques` |
number of blocks |

`cliquesize` |
size of each block |

`theta` |
magnitude of non-zeros |

### Details

This function generates a block diagonal positive definite
matrix with randomly-signed, non-zero elements. A shift is added to
the diagonal of the matrix so that its condition number equals
`p`

, the number of variables.

### Value

`Sigma` |
the covariance matrix |

`A` |
symmetric square root of |

`shift` |
how much the eigenvalues were shifted. See details. |

### Author(s)

Jacob Bien and Rob Tibshirani

### References

Bien, J., and Tibshirani, R. (2011), "Sparse Estimation of a Covariance Matrix," accepted for publication in Biometrika.