Is unbiasing estimators always justified ?

Jean-Michel Levy (1)

1 LPNHE - Laboratoire de Physique Nucléaire et de Hautes Énergies

It is argued that systematically unbiassing estimators is not always justified and can be counter-productive. On the contrary, a general rule is given to build a (slightly) biased but consistent estimator which is always closer to the value being estimated than the unbiased estimator from which one has started.

Mots clés: statistics; experimental data analysis

Domaines: Physique des Hautes Energies - Phénoménologie [hep-ph]

Jean-Michel Levy. Is unbiasing estimators always justified ?. 2006. ⟨hal-00022886v2⟩