| Type: | Package |
| Title: | Chauvenet-Type Boxplot |
| Version: | 1.0.0 |
| Description: | Provides a modified boxplot with a new fence coefficient determined by Lin et al. (2025). The traditional fence coefficient k=1.5 in Tukey's boxplot is replaced by a coefficient based on Chauvenet's criterion, as described in their formula (9). The new boxplot can be implemented in 'base R' with function chau_boxplot(), and in 'ggplot2' with function geom_chau_boxplot(). |
| Maintainer: | Tiejun Tong <tongt@hkbu.edu.hk> |
| URL: | https://tiejuntong.github.io/ChauBoxplot/ |
| License: | GPL-3 |
| Encoding: | UTF-8 |
| RoxygenNote: | 7.3.2 |
| Imports: | ggplot2, stats |
| Depends: | R (≥ 4.0.0) |
| Suggests: | testthat (≥ 3.0.0) |
| Config/testthat/edition: | 3 |
| NeedsCompilation: | no |
| Packaged: | 2025-05-17 10:08:53 UTC; TONGT |
| Author: | Hongmei Lin [aut], Riquan Zhang [aut], Tiejun Tong [aut, cre] |
| Repository: | CRAN |
| Date/Publication: | 2025-05-21 08:40:06 UTC |
ChauBoxplot: Chauvenet-type boxplot
Description
This package provides a modified boxplot with the new fence coefficient determined by Lin et al. (2025). Specifically, the fence coefficient k=1.5 in Tukey's boxplot has been replaced by the fence coefficient associated with Chauvenet's criterion in their formula (9). The Chauvenet-type boxplot can be implemented in base R with function chau_boxplot(), and in ggplot2 with function geom_chau_boxplot().
Author(s)
Maintainer: Tiejun Tong tongt@hkbu.edu.hk
Authors:
Hongmei Lin hmlin@suibe.edu.cn
Riquan Zhang rqzhang@suibe.edu.cn
References
Hongmei Lin, Riquan Zhang and Tiejun Tong (2025). When Tukey meets Chauvenet: a new boxplot criterion for outlier detection. Journal of Computational and Graphical Statistics, accepted.
See Also
Useful links:
Title: Chauvenet-type boxplot in base R
Description
This function can be operated the same way as boxplot() in base R, except that the fence coefficient k=1.5 in Tukey's boxplot is replaced by the fence coefficient associated with Chauvenet’s criterion. For details, please refer to formula (9) in Lin et al. (2025).
Usage
chau_boxplot(data, group_col = NULL, value_col = NULL, ...)
Arguments
data |
The data frame containing the data. |
group_col |
The column name for grouping data. |
value_col |
The column name for the values to plot. |
... |
Additional arguments passed to the plotting function. |
Value
A Chauvenet-type boxplot in base R.
References
Hongmei Lin, Riquan Zhang and Tiejun Tong (2025). When Tukey meets Chauvenet: a new boxplot criterion for outlier detection. Journal of Computational and Graphical Statistics, accepted.
Examples
# Example 1
chau_boxplot(c(rnorm(1000),5,6))
# Example 2
rate.senior <- c(4.96, 6.30, -5.38, 1.60, 7.24, 5.26, 2.55, 5.96, 3.96,
4.19, 1.88, 4.06, 4.75, 0, 0, 2.5, 2.87, 3.00)/100
chau_boxplot(rate.senior, notch=TRUE)
Title: Chauvenet-type boxplot in ggplot2
Description
This function can be operated the same way as geom_boxplot() in ggplot2, except that the fence coefficient k=1.5 in Tukey's boxplot is replaced by the fence coefficient associated with Chauvenet’s criterion. For details, please refer to formula (9) in Lin et al. (2025).
Usage
geom_chau_boxplot(
mapping = NULL,
data = NULL,
geom = "boxplot",
position = "dodge2",
na.rm = FALSE,
show.legend = NA,
inherit.aes = TRUE,
...
)
Arguments
mapping |
Aesthetic mappings created by 'aes()'. |
data |
The data to be displayed in this layer. |
geom |
The geometric object to use for display. |
position |
The position adjustment. |
na.rm |
Logical. Should missing values be removed? |
show.legend |
Logical. Should this layer be included in the legends? |
inherit.aes |
If FALSE, overrides the default aesthetics. |
... |
Other arguments passed to the layer. |
Value
A ggplot2 layer with the Chauvenet-type boxplot.
References
Hongmei Lin, Riquan Zhang and Tiejun Tong (2025). When Tukey meets Chauvenet: a new boxplot criterion for outlier detection. Journal of Computational and Graphical Statistics, accepted.
Examples
# Example 1
library(ggplot2)
rate.senior <- c(4.96, 6.30, -5.38, 1.60, 7.24, 5.26, 2.55, 5.96, 3.96,
4.19, 1.88, 4.06, 4.75, 0, 0, 2.5, 2.87, 3.00)/100
year <- 2007:2024
data.senior <- data.frame(x=year, y=rate.senior)
C.boxplot.senior <-
ggplot(data.senior, aes(y=rate.senior)) +
geom_chau_boxplot(fill="purple",width=3) +
theme(legend.position = "none") +
scale_x_discrete(breaks = NULL) +
ylim(-0.057,0.077) +
theme(plot.margin = unit(c(0, 0, 0, 0), "inches")) +
labs(title="C.boxplot", subtitle="Senior civil servants", x="", y="")
print(C.boxplot.senior)