Given a random sample of size `n`

from a continuous
distribution, then, with a confidence level of at least *γ*,
at least 100p percent of the population will be below the k*th* largest value in the
sample.

1 | ```
nptl(n , p = 0.95, gam = 0.95)
``` |

`n` |
the sample size |

`p` |
probability for Xp the 100p |

`gam` |
one-sided confidence level |

`k` |
index of the order statistic |

The maximum non-detect must be less than the k*th* largest
value.

E. L. Frome

Sommerville, P. N. (1958), "Tables for Obtaining Non-Parametric
Confidence Limits," *Annals of Mathematical Statistics*, 29, 599-601.

1 2 3 |

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

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