boxplot.stats: Box Plot Statistics

Description

This function is typically called by another function to gather the statistics necessary for producing box plots, but may be invoked separately.

Usage

boxplot.stats(x, coef = 1.5, do.conf = TRUE, do.out = TRUE)

Arguments

a numeric vector for which the boxplot will be constructed (NAs and NaNs are allowed and omitted).

coef

this determines how far the plot ‘whiskers’ extend out from the box. If coef is positive, the whiskers extend to the most extreme data point which is no more than coef times the length of the box away from the box. A value of zero causes the whiskers to extend to the data extremes (and no outliers be returned).

do.conf, do.out

logicals; if FALSE, the conf or out component respectively will be empty in the result.

Value

List with named components as follows:

stats

a vector of length 5, containing the extreme of the lower whisker, the lower ‘hinge’, the median, the upper ‘hinge’ and the extreme of the upper whisker.

the number of non-NA observations in the sample.

conf

the lower and upper extremes of the ‘notch’ (if(do.conf)). See the details.

out

the values of any data points which lie beyond the extremes of the whiskers (if(do.out)).

Note that $stats and $conf are sorted in increasing order, unlike S, and that $n and $out include any +- Inf values.

Details

The two ‘hinges’ are versions of the first and third quartile, i.e., close to quantile(x, c(1,3)/4). The hinges equal the quartiles for odd $n$ (where n <- length(x)) and differ for even $n$. Whereas the quartiles only equal observations for n %% 4 == 1 ($n\equiv 1 \bmod 4$), the hinges do so additionally for n %% 4 == 2 ($n\equiv 2 \bmod 4$), and are in the middle of two observations otherwise.

The notches (if requested) extend to +/-1.58 IQR/sqrt(n). This seems to be based on the same calculations as the formula with 1.57 in Chambers et al (1983, p.62), given in McGill et al (1978, p.16). They are based on asymptotic normality of the median and roughly equal sample sizes for the two medians being compared, and are said to be rather insensitive to the underlying distributions of the samples. The idea appears to be to give roughly a 95% confidence interval for the difference in two medians.

References

Tukey, J. W. (1977). Exploratory Data Analysis. Section 2C.

McGill, R., Tukey, J. W. and Larsen, W. A. (1978). Variations of box plots. The American Statistician, 32, 12--16. 10.2307/2683468.

Velleman, P. F. and Hoaglin, D. C. (1981). Applications, Basics and Computing of Exploratory Data Analysis. Duxbury Press.

Emerson, J. D and Strenio, J. (1983). Boxplots and batch comparison. Chapter 3 of Understanding Robust and Exploratory Data Analysis, eds. D. C. Hoaglin, F. Mosteller and J. W. Tukey. Wiley.

Chambers, J. M., Cleveland, W. S., Kleiner, B. and Tukey, P. A. (1983). Graphical Methods for Data Analysis. Wadsworth & Brooks/Cole.

Examples

Run this code

# NOT RUN {
require(stats)
x <- c(1:100, 1000)
(b1 <- boxplot.stats(x))
(b2 <- boxplot.stats(x, do.conf = FALSE, do.out = FALSE))
stopifnot(b1 $ stats == b2 $ stats) # do.out = FALSE is still robust
boxplot.stats(x, coef = 3, do.conf = FALSE)
## no outlier treatment:
boxplot.stats(x, coef = 0)

boxplot.stats(c(x, NA)) # slight change : n is 101
(r <- boxplot.stats(c(x, -1:1/0)))
stopifnot(r$out == c(1000, -Inf, Inf))

# }
# NOT RUN {
<!-- %% extended example (for the NG of Rdoc): -->
# }
# NOT RUN {
# }

Run the code above in your browser using DataCamp Workspace