Robust loss function derivatives — robust_loss

Computes the first derivative (psi function) of robust loss functions, evaluated at the given residuals. Used internally by ee_mean_robust() and ee_robust_regression().

This function mirrors robust_loss_functions() in Python delicatessen, so code translated from Python can keep its shape. There is no base R equivalent for these score functions, so this is the interface for them in deli as well.

Usage

robust_loss_functions(residuals, loss, k)

Arguments

residuals: Numeric vector of residuals.
loss: Character string specifying the loss function. One of: "huber", "tukey", "andrew", "hampel", "fair", "cauchy", "ullah", "welsch".
k: Numeric tuning constant. For "hampel", a length-3 vector c(a, b, c) where a < b < c.

Value

Numeric vector the same length as residuals.

Examples

r <- c(-5, -1, 0, 1, 5)
robust_loss_functions(r, "huber", k = 1.345)
#> [1] -1.345 -1.000  0.000  1.000  1.345
robust_loss_functions(r, "tukey", k = 4.685)
#> [1]  0.0000000 -0.9109563  0.0000000  0.9109563  0.0000000