Pure Error analysis of variance

Computes *Xa* where *X* is an *n by p* matrix, and *a* is
a random vector of length *p*.

1 2 3 4 | ```
randomLinComb(X, seed = NULL)
## S3 method for class 'lm'
randomLinComb(X, seed=NULL)
``` |

`X` |
An |

`seed` |
If not NULL, use this to initialize the random number generator |

The matrix *X* is first scaled and centered. *a* is computed to have random
uniform components on *(-1,1)*.

A vector of length *n*.

Sanford Weisberg, sandy@stat.umn.edu

1 | ```
randomLinComb(matrix(rnorm(9),ncol=3))
``` |

Questions? Problems? Suggestions? Tweet to @rdrrHQ or email at ian@mutexlabs.com.

Please suggest features or report bugs with the GitHub issue tracker.

All documentation is copyright its authors; we didn't write any of that.