This is an internal function to compute the process variance that replicates the minor error in Jaech's Fortran code. This allows comparing merror estimates to those shown in Jaech 1985.

1 | ```
process.var.mle.jaech.err(sigma2, s, beta.bars, N, n)
``` |

`sigma2` |
Estimated imprecisions for each method in an N vector. |

`s` |
Variance-covariance N x N matrix. |

`beta.bars` |
Estimates or hypothesized values for the N betas |

`N` |
No. of methods. |

`n` |
No. of items. |

See Jaech p. 186 equations 6.37 - 6.3.10. Jaech p. 288 line 2330 has s[i,j] instead of s[j,j]. Jaech p. 288 line 2410 omits "- 1/d2".

Estimated process variance but replicating minor error in Jaech's Fortran code.

Richard A. Bilonick

Jaech, J. L. (1985) *Statistical Analysis of Measurement Errors*. New York: Wiley.

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.