Construction of a geographically weighted nonparametric regression model fit test

Lilis Laome, I. Nyoman Budiantara*, Vita Ratnasari

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

1 Citation (Scopus)


One of the approach of Geographically Weighted Regression (GWR) models is the Geographically Weighted Nonparametric Regression (GWNR) has more parameters than the GWR model. Models with more parameters usually have better match values, which is an advantage, while models with fewer parameters have the advantage of being easier to use and interpret. However, a model with more parameters should be used if it is proven to be significantly superior. Therefore, the purpose of this study was to develop a hypothesis test of goodness of fit test for GWNR model. The goodness of fit test was performed for the real data. We found that the GWNR model was more suitable than the mixed nonparametric regression model.

Original languageEnglish
Article number102536
Publication statusPublished - Jun 2024


  • GWNR
  • Goodness of fit test
  • Infant Death
  • Poverty


Dive into the research topics of 'Construction of a geographically weighted nonparametric regression model fit test'. Together they form a unique fingerprint.

Cite this