Bivariate generalized Poisson regression (BGPR) is an extension of bivariate Poisson regression which deals overdipersion or underdispersion problem. This model gives global regression coefficients for all observations (locations) in the analysis. The BGPR model is then extended to take into account spatial heterogeneity, called geographically weighted bivariate generalized Poisson regression model, that yields varying regression coefficients locally. The regression model is applied to analyse factors affecting number of infant and maternal mortality in East Java, Indonesia.

Original languageEnglish
Pages (from-to)79-99
Number of pages21
JournalLetters in Spatial and Resource Sciences
Issue number1
Publication statusPublished - Apr 2021


  • Count data
  • Spatial analysis
  • Spatial heterogeneity


Dive into the research topics of 'Geographically weighted bivariate generalized Poisson regression: application to infant and maternal mortality data'. Together they form a unique fingerprint.

Cite this