radialplot.counts: Visualise point-counting data on a radial plot

In pvermees/provenance: Statistical Toolbox for Sedimentary Provenance Analysis

Visualise point-counting data on a radial plot

Description

Implementation of a graphical device developed by Rex Galbraith to display several estimates of the same quantity that have different standard errors.

Usage

## S3 method for class 'counts'
radialplot(
  x,
  num = 1,
  den = 2,
  from = NA,
  to = NA,
  t0 = NA,
  sigdig = 2,
  show.numbers = FALSE,
  pch = 21,
  levels = NA,
  clabel = "",
  bg = c("white", "red"),
  title = TRUE,
  ...
)

Arguments

x	an object of class counts
num	index or name of the numerator variable
den	index or name of the denominator variable
from	minimum limit of the radial scale
to	maximum limit of the radial scale
t0	central value
sigdig	the number of significant digits of the numerical values reported in the title of the graphical output.
show.numbers	boolean flag (TRUE to show sample numbers)
pch	plot character (default is a filled circle)
levels	a vector with additional values to be displayed as different background colours of the plot symbols.
clabel	label of the colour legend
bg	a vector of two background colours for the plot symbols. If levels=NA, then only the first colour is used. If levels is a vector of numbers, then bg is used to construct a colour ramp.
title	add a title to the plot?
...	additional arguments to the generic points function

Details

The radial plot (Galbraith, 1988, 1990) is a graphical device that was specifically designed to display heteroscedastic data, and is constructed as follows. Consider a set of dates \{t_1,...,t_i,...,t_n\} and uncertainties \{s[t_1],...,s[t_i],...,s[t_n]\}. Define z_i = z[t_i] to be a transformation of t_i (e.g., z_i = log[t_i]), and let s[z_i] be its propagated analytical uncertainty (i.e., s[z_i] = s[t_i]/t_i in the case of a logarithmic transformation). Create a scatterplot of (x_i,y_i) values, where x_i = 1/s[z_i] and y_i = (z_i-z_\circ)/s[z_i], where z_\circ is some reference value such as the mean. The slope of a line connecting the origin of this scatterplot with any of the (x_i,y_i)s is proportional to z_i and, hence, the date t_i. These dates can be more easily visualised by drawing a radial scale at some convenient distance from the origin and annotating it with labelled ticks at the appropriate angles. While the angular position of each data point represents the date, its horizontal distance from the origin is proportional to the precision. Imprecise measurements plot on the left hand side of the radial plot, whereas precise age determinations are found further towards the right. Thus, radial plots allow the observer to assess both the magnitude and the precision of quantitative data in one glance.

References

Galbraith, R.F., 1988. Graphical display of estimates having differing standard errors. Technometrics, 30(3), pp.271-281.

Galbraith, R.F., 1990. The radial plot: graphical assessment of spread in ages. International Journal of Radiation Applications and Instrumentation. Part D. Nuclear Tracks and Radiation Measurements, 17(3), pp.207-214.

Galbraith, R.F. and Laslett, G.M., 1993. Statistical models for mixed fission track ages. Nuclear Tracks and Radiation Measurements, 21(4), pp.459-470.

Examples

data(Namib)
radialplot(Namib$PT,components=c('Q','P'))

pvermees/provenance documentation built on Feb. 5, 2024, 4:50 a.m.