are {GeodRegr}

Approximate ARE of an M-type estimator to the least-squares estimator

Description

Approximate asymptotic relative efficiency (ARE) of an M-type estimator to the least-squares estimator given Gaussian errors, calculated using a tangent space approximation.

Usage

are(estimator, n, c = NULL)

Arguments

estimator

M-type estimator ('l2', 'l1', 'huber', or 'tukey').

n

Dimension of the manifold.

c

A positive multiplier, or a vector of them, of \sigma, the square root of the variance, used in the cutoff parameter for the 'huber' and 'tukey' estimators; should be NULL for the 'l2' or 'l1' estimators.

Value

Approximate ARE

Author(s)

Ha-Young Shin

References

Shin, H.-Y. and Oh H.-S. (2020). Robust Geodesic Regression. <arXiv:2007.04518>

See Also

are_nr.

Examples

are('l1', 10)