Description Usage Arguments Details Value References Examples
is_outlier
flags which observations fall outside a valid range of
values based on limits set by the user. Limits can be set in absolute
terms (the units of measurement), median absolute deviations, standard
deviations, or any combination of the above.
1 | is_outlier(measure, abs_lim = NULL, mad_lim = NULL, sd_lim = NULL)
|
measure |
A numeric vector. |
abs_lim |
Two-item numeric vector |
mad_lim |
Numeric value specifying the range of valid values in terms of median absolute deviations from the median. |
sd_lim |
Numeric value specifying the range of valid values in terms of standard deviations from the mean. |
If more than one type of limit is specified, is_outlier
will first
apply the absolute limits if given (so that values that are outright
impossible do not factor into the determination of the deviation statistics),
followed by the median-absolute-deviation (MAD) test and/or the
standard-deviation test.
The mad_lim
argument is evaluated using the double MAD, which
provides for robust identification of outliers even when the underlying
distribution is non-normal and/or asymmetric. See
Peter Rosenmai's blog post
for more information.
A logical vector indicating TRUE
if an observation is an
outlier.
Rosenmai P. 2013. Using the median absolute deviation to find outliers.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 | # Create example time series of 10 reaction times in ms with two trials that
# fall outside the bounds of validity:
set.seed(7)
rt <- c(0, 10000, rnorm(8, mean = 1000, sd = 250))
rt
# Check for trials that are less than 250 ms with no upper bound:
is_outlier(rt, abs_lim = c(250, Inf))
# Check for trials that are more than 3 standard deviations from the mean
is_outlier(rt, sd_lim = 3)
# Check for trials that are more than 3 median absolute deviations from the
# median
is_outlier(rt, mad_lim = 3)
# Check for trials that are less than 250 ms, more than 2500 ms, or more than
# 2.5 MADs from the median:
is_outlier(rt, abs_lim = c(250, 2500), mad_lim = 2.5)
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.