Weighting Factors

Helper function to transform uncertainty angles into weighting factors

Usage

weighting(
  x,
  method = c("linear-inverse", "inverse", "cosine", "none"),
  max.err = 90,
  norm = TRUE
)

Arguments

x

numeric. Uncertainty angle in degrees.

method

character. One of "linear-inverse" (the default), "inverse", "cosine", or "none" (no transformation).

max.err

numeric. The maximum expected error for x (90 by default).

norm

logical. Whether weights should be normalized so that the sum of the weights is 1. TRUE by default

Value

numeric

Details

Linear inverse: \(w = 1 - x/\sigma\), where \(\sigma\) is the maximum error expected for \(x\) (e.g. \(90^\circ\)).

Inverse: \(w = 1/x\)

Cosine: \(w = \cos{x}\)

Examples

x <- seq(0, 90, 1)

plot(x, weighting(x, "inverse"),
  col = 1, type = "l",
  xlab = "Uncertainty angle in degrees", ylab = "weight"
)
lines(x, weighting(x, "cosine"), col = 2)
lines(x, weighting(x, "linear-inverse"), col = 3)
legend("topright",
  col = 1:3, lty = 1,
  legend = c("inverse", "cosine", "linear-inverse")
)