homogeneityTests: 4 homogeneity tests as employed in the ECA&D project

In iki.dataclim: Consistency, Homogeneity and Summary Statistics of Climatological Data

Description

These functions implement the 4 homogeneity tests of climatological time series described in the Algorithm Theoretical Basis Document (ATBD) of the European Climate Assessment & Dataset project (ECA&D). These are the Standard Normal Homogeneity test (SNHtest), the BuisHand Range test (BHRtest), the PETtitt test (PETtest) and the VON Neumann ratio (VONtest).

Usage

SNHtest(X)
BHRtest(X)
PETtest(X)
VONtest(X)

Arguments

X	A numeric vector of equidistant observations. NAs are not allowed. At least 20 observations are required.

Details

The ATBD describes the test statistics for these 4 tests and provides critical values for significance levels of 1% and 5% for different sample sizes between 20 and 100. Significance is assessed conservatively, i.e. if the actual sample size lies between two of the provided sample sizes in the ATBD, we compare to the critical value of the smaller size. The Von Neumann ratio test (VONtest) only indicates whether the series has a break while the other tests also indicate the approximate location of the break.

Value

statistic	Test statistic
breakpoint	Index where the break occurs approximately.
significance	One of p5, p1 or NS for significant at 5%, at 1% or Not Significant, respectively.

Author(s)

Boris Orlowsky <[e-mail]>

References

The ATBD: http://www.ecad.eu/documents/atbd.pdf

See Also

evalHomogeneity

Examples

v <- c(rep(1, 10), rep(2, 10))
SNHtest(v)

Example output

$statistic
[1] 19

$breakpoint
[1] 10

$significance
[1] "p1"

iki.dataclim documentation built on May 2, 2019, 2:38 a.m.