cround: Rounding for Trigonometric Functions

In baddstats/spherstat: Spatial Point Pattern Analysis, Geometry and Trigonometry for the Sphere

Rounding for Trigonometric Functions

Description

Round the input for or output from trigonometric functions, to overcome some issues relating to rounding error.

Usage

cround(x)
sround(x)

Arguments

x	The number or numbers to be rounded. Can be a vector, matrix, numeric, list or data frame.

Details

The function cround rounds a number such that it is between -1 and 1. This overcomes rounding error whereby a number that has absolute value of 1 or very slightly less, is calculated by R to have absolute value of very slightly larger than 1, which can trigger errors if that calculated value is automatically passed to e.g. acos or asin. The function sround rounds a number that is extremely close to 0, to 0; this overcomes a minor calculation error in some R functions (e.g. trigonometric functions) and thus may trigger errors in some R functions (e.g. acos or asin, if the value being rounded is to be added to or subtracted from 1 or -1.

Value

The rounded values, in the same format as the input.

Warning

cround will round any number greater than 1 to 1; and any number less than -1 to -1. Care should be taken when using cround to ensure that its input are between (or close to one of) -1 and 1, that is that genuinely erroneous values are identified as such.

Author(s)

Tom Lawrence <email: [e-mail]>

Examples

  x <- c(-1-(10^17), -1+(10^-17), -0.5, 0, 0.5, 1-(10^-17), 1+(10^-17))
  cround(x)

  y <- c(-0.5, -10^-17, 10^-17, 0.5)
  sround(y)

baddstats/spherstat documentation built on Feb. 6, 2023, 1:45 a.m.