Analysis of correlated unit-Lindley data based on estimating equations

Silva D. V. V., AKDUR H. T. K., Paula G. A. A.

STATISTICAL METHODS AND APPLICATIONS, vol.32, no.5, pp.1477-1508, 2023 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 32 Issue: 5
  • Publication Date: 2023
  • Doi Number: 10.1007/s10260-023-00699-w
  • Journal Name: STATISTICAL METHODS AND APPLICATIONS
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus, ABI/INFORM, Applied Science & Technology Source, Business Source Elite, Business Source Premier, Computer & Applied Sciences, EconLit, Psycinfo, zbMATH
  • Page Numbers: pp.1477-1508
  • Keywords: Correlated data, Diagnostic procedures, Estimating equations, Unit-Lindley distribution
  • Gazi University Affiliated: Yes

Abstract

In this paper we derive estimating equations for modeling unbalanced correlated data sets in which the marginal distributions follow the one parameter unit-Lindley distributions with domain on the interval (0,1). A class of regressions models is proposed for modeling the location parameter and a reweighted iterative process is developed for the joint estimation of the regression coefficients and the correlation structure. Simulation studies are performed to assess the empirical properties of the derived estimators and diagnostic procedures, such as residual analysis and sensitivity studies based on conformal local influence are given. Finally, we analyze the proportion of people in households with inadequate water supply and sewage within federation units of Brazil by the procedures developed in the paper.