statsmodels.robust.norms.HuberT.rho

HuberT.rho(z)[source]

The robust criterion function for Huber’s t estimator.

\[\begin{split}\rho(z) = \begin{cases} \frac{1}{2} z^2 & \text{for } |z| \le t \\ t|z| - \frac{1}{2} t^2 & \text{for } |z| > t \end{cases}\end{split}\]
Parameters:
zarray_like

1d array

Returns:
rhondarray

The value of the robust criterion function.