knitr::opts_chunk$set( collapse = TRUE, comment = "#>", fig.path = "man/figures/README-", out.width = "100%" )
sensible summary statistics for big data
The goal of wodds
is to make the calculations of whisker odds (wodds) easy.
Wodds follow the same rules as letter-values, but with a different naming
system.
You can install the development version of wodds
from GitHub with:
# install.packages("devtools") devtools::install_github("alexhallam/wodds")
This is a basic example which shows you how to solve a common problem:
options(digits=1) library(wodds) library(knitr) set.seed(42) a <- rnorm(n = 1e4, 0, 1) df_wodds <- wodds::wodds(a) df_wodds
Outliers beyond the last wodd
are marked with O<value>
in ascending order. There should
rarely be more than 7 outliers when using wodds
.
df_wodds_and_outs <- wodds::wodds(a, include_outliers = TRUE) df_wodds_and_outs
Though not necessary it is possible to include tail area if additional communication or
teaching is needed. It is assumed that the wodd
should be explanatory enough
to not need to rely on tail_area
.
df_wodds_and_outs <- wodds::wodds(a, include_tail_area = TRUE) df_wodds_and_outs
An example with all options set to TRUE
.
df_wodds_and_outs <- wodds::wodds(a, include_depth = TRUE, include_tail_area = TRUE, include_outliers = TRUE) df_wodds_and_outs
A knitr::kable
example for publication.
knitr::kable(df_wodds_and_outs, align = 'c',digits = 3)
wodds::get_depth_from_n(n=15734L, alpha = 0.05)
wodds::get_n_from_depth(d = 11L)
Letter-Values are a fantastic tool! I think the naming could be improved. For this reason I introduce whisker odds (wodds) as an alternative naming system. My hypothesis is that with an alternative naming system the use of these descriptive statistics will be see more use. This is a rebranding of a what I think is a powerful modern statistical tool.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.