Boscovich's (1757) proposal to estimate the parameters of a linear model by minimizing the sum of absolute deviations subject to the constraint that the mean residual be zero is considered. The asymptotic theory of the estimator confirms a remark of Edgeworth who called it a "remarkable hybrid" between $\mathscr{l}_1$ and $\mathscr{l}_2$ methods.
Publié le : 1985-12-14
Classification:
Least absolute error estimators,
linear models,
asymptotic theory,
62F12,
62J07
@article{1176349759,
author = {Koenker, Roger and Bassett, Gilbert},
title = {On Boscovich's Estimator},
journal = {Ann. Statist.},
volume = {13},
number = {1},
year = {1985},
pages = { 1625-1628},
language = {en},
url = {http://dml.mathdoc.fr/item/1176349759}
}
Koenker, Roger; Bassett, Gilbert. On Boscovich's Estimator. Ann. Statist., Tome 13 (1985) no. 1, pp. 1625-1628. http://gdmltest.u-ga.fr/item/1176349759/