A statistical test used to determine whether a set of angular or circular data points (such as times of day, compass directions, or degrees) are spread out evenly around a circle or if they cluster in some way.
Arguments
- x
numeric vector. Values in degrees
- alpha
Significance level of the test ranging between 0 and 1.
- axial
logical. Whether the data are axial, i.e. \(\pi\)-periodical (
TRUE, the default) or directional, i.e. \(2 \pi\)-periodical (FALSE). In case of axial data, the angles will be doubled for the test.- quiet
logical. Prints the test's decision.
Value
A list with class "htest" containing the following components:
statisticthe value of the modified V statistic
parameterthe number of data
p.valuethe p-value for the test
estimatethe D+ and D- values
alternativea character string describing the alternative hypothesis.
methoda character string indicating what type of Kuiper test was performed.
data.namea character string giving the name(s) of the data.
criticalthe critical value at the significance level
alphathe significance level
rejectlogical. whether to reject the null-hypothesis based on the significance level.
Details
The Null Hypothesis (\(H_0\)): The data are distributed completely uniformly (randomly and evenly) around the circle.
The Alternative Hypothesis (\(H_1\)): The data are not uniform and show a preference, clustering, or pattern somewhere on the circle.
The Test Statistic (V or \(D^{+} + D^{-}\)): It measures the greatest positive and negative differences between your data's empirical cumulative distribution and a theoretical uniform distribution.
Interpreting the Results
High Test Statistic / Low p-value (\(p < \alpha\), typically 0.05): You reject the null hypothesis. This means your data are not uniform; they have a significant preferred direction, grouping, or non-random pattern on the circle.
Low Test Statistic / High p-value (\(p \ge 0.05\)): You fail to reject the null hypothesis. There is no strong evidence to say the data are different from a flat, uniform distribution. The points appear random across the circle.
Note
Kuiper's test statistic is a rotation-invariant Kolmogorov-type test statistic. The modified statistic of Stephens (1970) is compared with the asymptotic Kuiper distribution.
See also
Other One-Sample-Tests:
hodges_ajne_test(),
norm_chisq(),
rao_spacing_test(),
rayleigh-test,
watson_test(),
weighted-rayleigh-test,
weighted_rayleigh_moore()
Examples
# Example data from Mardia and Jupp (1999), pp. 93
kuiper_test(homing, alpha = .05)
#> Reject Null Hypothesis
#>
#> Kuiper's test of uniformity
#>
#> data: homing
#> V = 2.2621, n = 10, p-value = 0.001398
#> alternative hypothesis: non-uniform distribution
#> sample estimates:
#> D+ D-
#> 0.3611111 0.3055556
#>
# San Andreas Fault Data:
data(san_andreas)
data("nuvel1")
PoR <- subset(nuvel1, nuvel1$plate.rot == "na")
sa.por <- PoR_shmax(san_andreas, PoR, "right")
kuiper_test(sa.por$azi.PoR, alpha = .05)
#> Reject Null Hypothesis
#>
#> Kuiper's test of uniformity
#>
#> data: sa.por$azi.PoR
#> V = 16.605, n = 1126, p-value < 2.2e-16
#> alternative hypothesis: non-uniform distribution
#> sample estimates:
#> D+ D-
#> 0.007774912 0.484679833
#>
