Skip to main navigation Skip to search Skip to main content

Spatial scalar-on-function quantile regression model

  • Muge Mutis*
  • , Ufuk Beyaztas
  • , Filiz Karaman
  • , Han Lin Shang
  • *Corresponding author for this work

Research output: Contribution to journal › Article › peer-review

Abstract

This paper introduces a novel spatial scalar-on-function (SoF) quantile regression model that extends classical SoF models to account for spatial dependence and heterogeneous conditional distributions. The proposed model incorporates spatial autocorrelation through a spatially lagged response and characterises the entire conditional distribution of a scalar outcome given a functional predictor. To address endogeneity induced by the spatial lag term, we develop two robust estimation procedures based on instrumental variable (IV) strategies. The √𝑛-consistency and asymptotic normality of the proposed estimators are established under mild regularity conditions. We demonstrate through extensive Monte Carlo simulations that the proposed estimators outperform existing mean-based and robust alternatives, particularly in settings with strong spatial dependence and outlier contamination. We apply our method to high-resolution environmental data from the Lombardy region in Italy, using daily ozone trajectories to predict daily mean concentrations of particulate matter with a diameter of less than 2.5 micrometers. The empirical results confirm the superiority of our approach in predictive accuracy, robustness and interpretability at various quantile levels. Our method has been implemented in the ssofqrm image package.
Original languageEnglish
Article numbere70062
Pages (from-to)1-28
Number of pages28
JournalAustralian and New Zealand Journal of Statistics
Volume68
Issue number3
DOIs
Publication statusPublished - Sept 2026

Keywords

  • functional principal component analysis
  • PM2.5 concentration
  • quantile regression
  • spatial autoregressive model
  • spatial dependence

Fingerprint

Dive into the research topics of 'Spatial scalar-on-function quantile regression model'. Together they form a unique fingerprint.

Cite this