vareval: A data assessment and screening tool

In johnlpearce/sommix: Tools to Apply Self-Organizing Maps

Description

Performs a generalized evaluation of the data in order to examine standardized distributions of data via boxplots, variable specific missing-value patterns, and variable specific zero-value patterns. A table of summary statistics is also provided.

Usage

vareval(x, labsize = 1, symsize = 2, colormod = NULL, varlabs = NULL)

Arguments

x	is a data frame object
labsize	specifies cex value of labels on plots default is 0.75
symsize	specifies cex value of symbols on plots default is 2
colormod	can be used to customize plot color schemes. The default uses terrain palletes from grDevices
varlabs	specifies x-axis variable labels

Details

An important first step in the evaluation of multivariate data sets is to examine candidate variables individually. This assessment and screening tool is designed to assist understanding individual data distributions and patterning with problematic features such as missing and zero values.

Value

a dataframe consisting of summary statistics

• VARIABLE Name of variable

• TYPE variable class

• N number of observations

• NAs number of missing values

• ZEROs number of zero values

• MEAN mean value

• SD standard deviation

• MIN minimum value

• Q1 lower quartile

• MED median value

• Q3 upper quartile

• MAX maximum value

• IqR interquartile range

• CoV coefficient of variation, i.e., SD/MEAN

• CVrank ranking of CoV values

Examples

#NIEHS Mixtures Workshop dataset1
data(dataset1)
vareval(dataset1)

johnlpearce/sommix documentation built on Jan. 7, 2021, 11:38 p.m.