Geographically Weighted Regression (GWR) model is a powerful tool for analyzing spatial patterns in data. However, the standard form of a spatial model that uses a single bandwidth calibration may be unrealistic because the response-predictor relationship may be either linear or nonlinear. To address this issue, the Multiscale GWR (MSGWR) model offers improved model performance by employing Generalized Additive Model (GAM) with varying bandwidth or smoothing function for each covariate in the model. This research aims to analyze the Percentage of Poor Population (PPP) on Java Island in 2022 using the geospatial models and related socioeconomic and demographic attributes, such as Open Unemployment Rate, Human Development Index, Labor Force Participation Rate, and GRDP Per capita to identify the best model in explaining the spatial pattern and to find out the determinant of PPP on Java Island in 2022. This study uses secondary data from Statistics Indonesia. The findings reveal that the MSGWR model provides the highest R2 and smallest AICc value compared to single bandwidth models, specifically the GWR and MXGWR models. Furthermore, the MSGWR model indicates that HDI has a significant negative effect on PPP, whereas LFPR has a significant positive effect on PPP across all districts in Java Island in 2022.
Jaringan kolaborasi para penulis
4 penulis ·
6 tautan kepenulisan bersama
· 3 lintas organisasi
Arahkan kursor atau ketuk sebuah nama untuk menyorot rekan penulisnya. Klik dua kali untuk membuka halaman penulis.
Politeknik Statistika STIS
Binus University
Organisasi berbeda
Organisasi sama
Afiliasi tidak diketahui
Garis lebih tebal = lebih banyak artikel bersama · lingkaran lebih besar = lebih banyak artikel
Afiliasi berasal dari OpenAlex; metadata OAI-PMH jurnal tidak memuatnya. Penulis tanpa ID OpenAlex dicocokkan berdasarkan nama, sehingga dua orang bisa tergabung atau satu orang bisa terpisah.