Child Height-for-Age and Remoteness in Timor-Leste: A Lasso Quantile Regression Analysis

Melbourne Institute Working Paper No. 05/26

Date: May 2026

Author(s):

Diana Contreras Suárez
Silvia Espinosa

Abstract

This paper analyses the factors influencing height-for-age Z-scores (HAZ) of children in Timor-Leste, a country with one of the highest stunting prevalences in Southeast Asia and globally (47.1%). We use lasso quantile regression to select the most relevant variables from a rich dataset of potential determinants of HAZ, and estimate the effects of these determinants at four quantiles of the HAZ distribution: severely stunted, stunted, mildly-undernourished, and well-nourished children. The results show substantial heterogeneity in the effects of the determinants of HAZ across its distribution, and according to the household’s proximity to the regional capital. These findings provide empirical evidence that nutrition interventions should be differentiated based on both children’s nutritional status and geographic location.

To obtain a copy of this paper, please email the author.
Request Paper