GGBK: Cohn Approximation for New Generalized Grubbs-Beck Critical...

In MGBT: Multiple Grubbs-Beck Low-Outlier Test

Description

Compute Cohn's approximation of critical values at the 10-percent significance level for the new generalized Grubbs–Beck test.

Usage

GGBK(n)

Arguments

n	The number of observations.

Value

The critical value of the test was to be returned, but a stop() is issued instead because there is a problem with the function's call of CritK (see Note).

Note

In TAC sources, GGBK is the consumer of two global scope functions fw() and fw1(). These should be defined within the function to keep the scope local as they are unneeded anywhere else in TAC sources, and these thus have local scope in the implementation for the MGBT package.

A BUG FIX NEEDED—Note that TAC has a problem in sources in that this function is incomplete. The function CritK is the issue, that function requires three arguments and appears to work (see Examples under CritK), but TAC code passes four in the context of GGBK. At present (packaging of MGBT), it is not known if the word “generalized” in this test has the same meaning as “multiple” in the Multiple Grubbs–Beck Test. Also in TAC sources, it is as yet unclear what the “new” in the title of this function means.

Author(s)

W.H. Asquith consulting T.A. Cohn sources

Source

LowOutliers_jfe(R).txt, LowOutliers_wha(R).txt, not P3_089(R).txt—Named GGBK

References

Cohn, T.A., 2013–2016, Personal communication of original R source code: U.S. Geological Survey, Reston, Va.

See Also

critK10, CritK

Examples

## Not run: 
GGBK(34) # but presently the function does not work 
## End(Not run)

MGBT documentation built on July 22, 2021, 1:06 a.m.