Skip to contents

Descriptive statistics for calendar date variables

Source: R/Desc.Date.R
Desc.Date.Rd

Computes a structured descriptive summary for objects of class "Date". The description focuses on time-axis characteristics (range, span, coverage, quantiles) and distributional structure over weekdays and months.

Usage

# S3 method for class 'Date'
desc(
  x,
  main = NULL,
  plotit = NULL,
  verbose = NULL,
  wprobs = rep(1/7, 7),
  mprobs = rep(1/12, 12),
  ...
)

Arguments

x

a dichotomous numeric, integer, factor, character, or logical vector

main

character string, NULL, or NA, defining the main title. By default (main = NULL) the title will be composed as: (<class(es)>). If NA, no title is printed.

plotit

logical. Should a plot be created? The plot type depends on the classes of the variables. Default can be defined by the option plotit, if it does not exist then it's set to TRUE.

verbose

integer controlling verbosity of table output. One of 1 (minimal), 2 (default), 3 (extensive). Applies to tables only.

wprobs

numeric vector of length 7 specifying expected probabilities for weekdays (Monday to Sunday). The default is a uniform distribution rep(1/7, 7).

mprobs

numeric vector of length 12 specifying expected probabilities for months (January to December). If NULL (default), probabilities proportional to the number of days per month in a non-leap year are used.

...

further arguments passed to methods

Value

an object of class c("Desc.Date", "Desc") with components:

core

time-axis statistics

weekday

observed and expected weekday counts, standardized residuals, and p-value

month

observed and expected month counts, standardized residuals, and p-value

sentinel

heuristic data-quality diagnostics

meta

metadata

Details

In addition to core time-axis statistics, observed and expected frequencies for weekdays and months are calculated together with standardized residuals and chi-square p-values. The function also performs heuristic detection of suspicious sentinel dates (e.g., extreme future or implausibly early values) to highlight potential data-quality issues.

The core time-axis summary includes:

  • Number of observations and missing values

  • Minimum and maximum date

  • Span in days (max - min)

  • Number of unique observed days

  • Coverage: proportion of observed days relative to the total number of calendar days within the observed range

  • Fundamental quantiles (5\

  • Interquartile range (IQR) in days

Weekday and month distributions are compared to their expected probabilities using chi-square goodness-of-fit tests.

Standardized residuals are defined as $$(Observed - Expected) / sqrt(Expected)$$. They describe the magnitude and direction of deviation from the expected distribution.

Sentinel detection is based on simple heuristics such as extremely large future dates or implausibly early calendar dates. It is meant as a diagnostic aid rather than a formal validation procedure.

See also

desc(), print.Desc.Date()

Other desc: desc(), desc.factor(), desc.nn, desc.nq, desc.numeric(), desc.qn, desc.qq, desc.table(), desc.ts()