Confidence Interval around the Mean Direction of Circular Data after Fisher (1993)
Source:R/statistics.R
confidence_interval_fisher.Rd
For large samples (n >=25
) i performs are parametric estimate based on
sample_circular_dispersion()
. For smaller size samples, it returns a
bootstrap estimate.
Usage
confidence_interval_fisher(
x,
conf.level = 0.95,
w = NULL,
axial = TRUE,
na.rm = TRUE,
boot = FALSE,
R = 1000L,
quiet = FALSE
)
Arguments
- x
numeric vector. Values in degrees.
- conf.level
Level of confidence: \((1 - \alpha \%)/100\). (
0.95
by default).- w
(optional) Weights. A vector of positive numbers and of the same length as
x
.- axial
logical. Whether the data are axial, i.e. pi-periodical (
TRUE
, the default) or directional, i.e. \(2 \pi\)-periodical (FALSE
).- na.rm
logical value indicating whether
NA
values inx
should be stripped before the computation proceeds.- boot
logical. Force bootstrap estimation
- R
integer. number of bootstrap replicates
- quiet
logical. Prints the used estimation (parametric or bootstrap).
Examples
# Example data from Davis (1986), pp. 316
finland_stria <- c(
23, 27, 53, 58, 64, 83, 85, 88, 93, 99, 100, 105, 113,
113, 114, 117, 121, 123, 125, 126, 126, 126, 127, 127, 128, 128, 129, 132,
132, 132, 134, 135, 137, 144, 145, 145, 146, 153, 155, 155, 155, 157, 163,
165, 171, 172, 179, 181, 186, 190, 212
)
confidence_interval_fisher(finland_stria, axial = FALSE)
#> Parametric estimate
#> $mu
#> [1] 129.1903
#>
#> $conf.angle
#> [1] 10.20306
#>
#> $conf.interval
#> [1] 118.9872 139.3933
#>
confidence_interval_fisher(finland_stria, axial = FALSE, boot = TRUE)
#> Bootstrap estimate based on 1000 replicates
#> $mu
#> [1] 129.1638
#>
#> $conf.angle
#> [1] 6.108974
#>
#> $conf.interval
#> [1] 122.9056 135.1285
#>
data(san_andreas)
data("nuvel1")
PoR <- subset(nuvel1, nuvel1$plate.rot == "na")
sa.por <- PoR_shmax(san_andreas, PoR, "right")
confidence_interval_fisher(sa.por$azi.PoR, w = 1 / san_andreas$unc)
#> Parametric estimate
#> $mu
#> [1] 142.403
#>
#> $conf.angle
#> [1] 3.763676
#>
#> $conf.interval
#> [1] 138.6393 146.1667
#>
confidence_interval_fisher(sa.por$azi.PoR, w = 1 / san_andreas$unc, boot = TRUE)
#> Bootstrap estimate based on 1000 replicates
#> $mu
#> [1] 139.3926
#>
#> $conf.angle
#> [1] 1.040634
#>
#> $conf.interval
#> [1] 138.4089 140.4930
#>