REGIONAL CLASSIFICATION BASED ON MATERNAL MORTALITY RATE USING A ROBUST SEMIPARAMETRIC GEOGRAPHICALLY WEIGHTED POISSON REGRESSION MODEL

Abstract

The Semiparametric Geographically Weighted Poisson Regression (SGWPR) model advances the GWPR model, combining local and global parameters relative to location. Outliers are sometimes encountered when analyzing data using the GWPR model. These outliers can be identified as they differ significantly from other sample points. Outliers can impact the estimation results, leading to biased parameter estimates. One approach to addressing outliers is the robust M method. This study aims to classify regions based on the parameter estimates of the robust SGWPR model applied to maternal mortality rate data in East Java Province using Tukey's Bisquare weighting. The outcome of this research is the classification of regions based on significant factors influencing maternal mortality rates in East Java Province in 2021.

Access to Document

Other files and links

• Link to publication in Scopus
• Open Access Version Available

Fingerprint