Book chapter

Bias Calibration for Robust Estimation in Small Areas

PublisherCham : Springer International Publishing
Publication date2023-11-26
First online date2022-11-26

It is well known that the existence of outliers in a sample can significantly

affect the estimation of population parameters. Intuition suggests that this is even

more the case in the context of small area estimation. If influential, outliers may

heavily affect parameter estimates for areas in which they occur, especially when the

domain-sample size is tiny. An obvious remedy is to use robust estimators but with

the drawback of a potential bias. We compare different approaches, including some

new ones, for bias calibration in this context. Among other findings, the simulations

indicate that the new proposals can lead to more efficient estimators compared to

existing methods. We conclude the study applying our estimators to obtain Gini

coefficients in labour market areas of the Tuscany region of Italy. The new methods

reveal a different picture than existing methods. We extend our ideas to predictions

for non-sampled areas.

  • AsymmetricHuberfunction
  • Non-linearpopulationparameters
  • Robust estimation
  • Robust prediction
  • Small area estimation
Citation (ISO format)
RANJBAR, Setareh, RONCHETTI, Elvezio, SPERLICH, Stefan Andréas. Bias Calibration for Robust Estimation in Small Areas. In: Robust and Multivariate Statistical Methods. Cham : Springer International Publishing, 2023. p. 365–394. doi: 10.1007/978-3-031-22687-8_17
Main files (1)
Book chapter (Published version)

Technical informations

Creation01/11/2024 7:36:18 PM
First validation01/15/2024 7:28:34 AM
Update time01/15/2024 7:28:34 AM
Status update01/15/2024 7:28:34 AM
Last indexation05/06/2024 5:44:52 PM
All rights reserved by Archive ouverte UNIGE and the University of GenevaunigeBlack