Small area estimation (SAE) aims to allow efficient estimation of population characteristics of domains with a small sample size that produce unreliable estimates. In the last decade there has been a rising interest in poverty estimation where the ELL (a simulation-based synthetic poverty mapping methodology) approach is the de-facto industry standard for small area estimation applied to poverty assessment. Alternatives to the ELL approach take into account area and unit heterogeneity, but implicitly assume homogeneity for clusters within area level (e.g. PSU). We propose a two-level M-quantile linear model that should be able to capture variability at area/domain and cluster level. The model in its simplest form can mimic a two-nested-error model. Using a Monte Carlo approach we can obtain poverty estimates via the two- level M-quantile model. Performance of poverty estimates will be shown by means of Monte Carlo simulation where we tried to build a realistic poverty estimation scenario.

Two-level M-quantile model for poverty estimation

MARCHETTI, STEFANO;SALVATI, NICOLA
2015-01-01

Abstract

Small area estimation (SAE) aims to allow efficient estimation of population characteristics of domains with a small sample size that produce unreliable estimates. In the last decade there has been a rising interest in poverty estimation where the ELL (a simulation-based synthetic poverty mapping methodology) approach is the de-facto industry standard for small area estimation applied to poverty assessment. Alternatives to the ELL approach take into account area and unit heterogeneity, but implicitly assume homogeneity for clusters within area level (e.g. PSU). We propose a two-level M-quantile linear model that should be able to capture variability at area/domain and cluster level. The model in its simplest form can mimic a two-nested-error model. Using a Monte Carlo approach we can obtain poverty estimates via the two- level M-quantile model. Performance of poverty estimates will be shown by means of Monte Carlo simulation where we tried to build a realistic poverty estimation scenario.
2015
9789963222704
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11568/766108
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact